Handbook of Extractive Metallurgy

Reprint from:

Handbook of Extractive Metallurgy Edited by Fathi Habashi

WILEY-VCH Weinheim • Chichester • New York • Toronto • Brisbane • Singapore ©VCH Verlagsgesellschaft mbH, Weinheim, Federal Republic of Germany, 1997

17 Mercury MATTHIAS SIMON (§§ 1 7 . 1 - 1 7 . 1 0 , 1 7 . 1 3 - 1 7 . 1 4 ) ; PETER JÖNK (RETIRED) (§ 1 7 . 1 1 ) ; GABRIELE WÜHL-COUTURIER (§ 1 7 . 1 2 ) ; M A X DAUNDERER (§ 17.14)

17.1 Introduction 17.2 Properties 17.2.1 17.2.2

17.3

Physical Properties C h e m i c a l Properties

Resources and Raw Materials

17.3.1 17.3.2

17.4 17.4.1 17.4.1.1 17.4.1.2 17.4.1.3 17.4.2 17.4.3 17.4.4

Deposits Secondary Sources

Production

17.6 891 17.7 892 17.8 893 17.9 893 895 17.10 8 9 5 17.11

Extraction from Primary Sources 895 Beneficiation 895 Processing to Metallic Mercury 895 Furnace Systems 896 Extraction from Secondary Sources 897 Condensation o f Mercury from Furnace Off-Gas 898 Treatment o f t h e Stupp 900

17.5 Environmental Protection 1 7 . 5 . 1 Natural Distribution o f M e r c u r y 17.5.1.1 Mercury in Soil, Plants, and Animals 17.5.1.2 Mercury in Food 1 7 . 5 . 2 M e r c u r y Emissions 17.5.2.1 Gas Purification

905

17.5.2.2 Water Purification

891 891

Quality Specifications Chemical Analysis Storage and Transportation Uses Mercury Alloys Compounds

1 7 . 1 1 . 1 Mercury

Chalconides

910

1 7 . 1 1 . 2 Mercury

Halides

911

1 7 . 1 1 . 3 Mercury Pseudohalides

913

1 7 . 1 1 . 4 Acetates, Nitrates, Sulfates

913

1 7 . 1 1 . 5 Mercury-Nitrogen C o m p o u n d s

914

1 7 . 1 1 . 6 Analysis, Storage, and Transportation; Protective Measures 914

901

17.12 Selected Organic Compounds

901 901 902 904 904

1 7 . 1 3 Economic Aspects 17.14 Toxicology and Occupational Health 17.15 References

17.1 Introduction Mercury (also called quicksilver), symbol Hg from the Greek νδραρ = water and αργνρος = silver, was known as early as 1000 B . C . because of its liquid state at room tem­ perature (mp -38.89 °C). The discovery in 1938 of 1 kg of the metal in 2500-year-old sand layers on the eastern coast of Greece in­ dicates that mercury was used in the extraction of gold at an early date [1]. Mercury was men­ tioned about 200 B . C . in India as well as in China (Han dynasty). As early as 1556, GEORGES BAUER, known

as AGRICOLA, reported five different methods for extracting mercury from its ores [2]. He also realized that mercury vapors are heavier than air and that they could therefore conve­ niently be trapped in condensers beneath the reaction vessel.

906 906 907 907 908 910

915 915 917 920

Because of the considerable vapor pressure of mercury even at room temperature and the toxicity of its vapors, particularly safe and re­ liable methods must be used in the extraction of mercury to avoid releasing even the slight­ est trace of the metal into the environment. The problem of gas and water purification is therefore particularly important.

17.2 Properties 17.2.1

Physical Properties

Mercury is a silvery-white, shiny metal, which is liquid at room temperature. The most important physical properties are listed below: Isotope masses (ordered according 2 0 2 , 2 0 0 , 1 9 9 , 2 0 1 t o decreasing abundance) 198 204, 196 - 3 8 . 8 9 °C

mp bp ( 1 0 1 . 3

kPa)

357.3

°C

Density (0 °C) 13.5956 g/cm 3 Specific heat capacity c p (0 °C) 0.1397 J g - l K - 1

892

Handbook

Heat of fusion

11.807 J/g

Heat of evaporation (357.3 °C)

59.453 kJ/mol

Thermal conductivity (17 °C)

0.082 W cm K

-1

-4

1.826 x 10

K

-4

-1

-1 -1

Electrical conductivity (0 °C)

1.063 x 10

Crystal structure

rhombohedral

mΩ mm

-2

1.685 mPa •s

Viscosity (0 °C)

-5

480.3 x 10 N/cm

Surface tension tcritt

1450 °C

pcritt Critical density Evaporation number (25 °C)

105.5 MPa 5 g/cm

3 -1

0.085 mg K cm

-2

Mercury has a relatively high vapor pres­ sure, even at room temperature. Saturation va­ por pressures at 0-100 °C are listed in Table 17.1 (corresponding to a specified mercury content in air). The temperature dependence of the density of mercury is given in Table 17.2. Mercury vapor is excited to a state of lumi­ nescence by electrical discharge (mercury va­ por lamps). Ultraviolet radiation is released primarily, it can be used to start and to pro­ mote chemical reactions. Table 17.1: Saturation vapor pressure of mercury at dif­ ferent temperatures. p, Pa

Hg content in air, g/m3

0

0.026

0.00238

10

0.070

0.00604

20

0.170

0.01406

30

0.390

0.03144

100

36.841

2.40400

t, °C

The dynamic viscosity η of mercury (Table 17.3) is of the same order of magnitude as that of water [3]. Some other metals, such as gold, silver, and zinc, dissolve readily in mercury to form amalgams. The solubility of mercury in water is strongly temperature dependent [4]. At room temperature, the solubility is ca. 60 μg/L; at 50 °C, ca. 250 μg/L; and at 90 °C, ca. 1100 μg/L. At low temperature, the addition of sodium chloride reduces the water solubil­ ity.

17.2.2

Mercury exists in the oxidation states 0, 1+, and 2+; monovalent mercury is found only in Hg-Hg bonds. The redox potentials E 0 at 298.15 K and 101.325 kPa relative to the stan­ dard hydrogen electrode are as follows [3]: Hg2+ + 2e2Hg Hg

2+

2+

+ 2e

2

-38.89 0 50 100 150

13.6902 13.5956 13.4733 13.3524 13.2327

t, °C Density, g/cm3 200 250 300 350

13.1139 12.9957 12.8778 12.7640

Table 17.3: Temperature dependence of the dynamic vis­ cosity of mercury.

3

Density, g/cm 3

-20 -10 0 10 20 30 40

1.855 1.764 1.685 1.615 1.554 1.499 1.450

t, °C Density, g/cm3 60 80 100 200 300 340

-

Hg

1.367 1.298 1.240 1.052 0.950 0.921

2+

+0.920 V

2

2Hg

-

2

+0.797 V 2Hg + 2CH COO 3

+0.511 V

The standard potentials show that mercury is a relatively noble metal. Metallic mercury dissolves in nitric acid, aqua regia, warm concentrated hydrochloric acid and sulfuric acid. It is sparingly soluble in dilute HC1, HBr, and HI as well as in cold sulfuric acid. Most of its chemical compounds have densities of 5-9 g/cm . The oxide of mercury (HgO) decomposes at 400-500 °C. This effect is utilized in the extraction of mercury from oxidic sources. Mercury forms monovalent and divalent compounds with the halogens fluorine, chlorine, bromine, and iodine. It also forms monovalent and divalent compounds with sulfur. 3

t, °C

+0.851 V

Hg

+ 2e

2

Density, g/cm3

Chemical Properties

Hg (CH COO) + 2e-

-

Table 17.2: Density of mercury as a function of tempera­ ture. t, °C

Metallurgy

The surface tension of mercury is ca. six times greater than that of water, which is the reason for its poor wettability by water.

Thermal expansion coefficient β (0-100 °C)

of Extractive

Mercury

893

17.3 Resources and Raw Materials 17.3.1

Deposits

All known mercury ores are relatively lowgrade ores, the average mercury content being ca. 1 %. Because mercury ores lie close to the earth's surface, the mining depth is ca. 800 m at most. The overwhelming proportion of mercury has always been produced in Europe; some of the most important deposits are listed below. Over the last ten years, the former Soviet Union, Spain, China, Algeria, Turkey, and the United States have accounted for ca. 90 % of world mining production. Spain. The deposits at Almadén have been mined for more than 2000 years. Almadén is situated in the southwest of the province of Ciudad Real in New Castile, about 200 km from Madrid, on the northern edge of the Sierra Morena. The ore-bearing deposits are porous sedimentary rocks (sandstone, bituminous shale, Silurian quartzite) that contain mercury sulfide (cinnabar). Needles of coarsely crystalline cinnabar as well as metallic mercury in lenticular ore-bearing bodies are also found over a roughly 20-km stretch in the valley of the Valdeazogues river. The ore is extracted in several mines (San Pedro y San Diego, San Francisco, San Nicolas). The richest strata contained 12-14% mercury at a depth of 170-200 m. Mining is carried out today at a depth of 500 m; 1 t of ore yields about one flask of mercury (= 34.473 kg of mercury, corresponding to a 3.5% mercury content in the ore). This deposit was probably known to the Celts and Phoenicians, and was mined by the Romans from 150 B.C. These mines were subsequently worked by the Moors, then by the orders of knights, by the Spanish royal house, and from 1525-1645 by the Fuggers. Since then the mines have been under state management. From 1449 (when production records start) to the present, ca. 300 000 t of mercury has been mined.

In 1988 Almadén produced ca. 1380 t of mercury. The ore-bearing body has an average mercury content of 5%. Because of the low price of mercury the plant operates for only a few months of the year. The operating company, Minas de Almadén y Arrayanes SA, is state-owned [5,6]. Italy. A roughly 25-km-wide, 50-km-long belt running from Monte Amiata to the coast in southern Tuscany contains many closed mines as well as some that are still operating. Although these deposits had been mined by the Etruscans, they were not intensively worked by the Romans, to the benefit of the Spanish mines at Almadén. The Siele mine recommenced operation only in 1846, and the Abbadia San Salvatore mine, at present still the largest, in 1898. Mercury occurs as cinnabar and metacinnabarite in pyrites, marcasite, antimonite, and realgar, mainly in Eocene sediments (shale, sandstone, marl, limestone) under a covering layer of trachyte. The gangue consists largely of clay or dolomite. The ores contain on average 0.2-0.8% mercury. Mercury extraction is organized by the Monte Amiata works, its latent production capacity being up to 2000 t/a. No mercury has been mined since 1983. Former Yugoslavia. The third largest mercury deposits in Europe are situated in Idrija in Slovenia, about 40 km west of Ljubljana. These deposits have been worked since 1497. The mines were under Austrian ownership from 1580 to 1918, then under Italian ownership, and finally reverted to Yugoslavia after World War II. The tectonics of this region are complicated. Shell marl and shale are impregnated with cinnabar and native mercury, and typically contain admixtures of pyrites, gypsum, and bitumen. The ore contains ca. 0.5% mercury. Idrija has a capacity of about 600 t/a. About 60 t mercury was mined in 1986. Additional deposits in Yugoslavia are located at Maškara in Croatia, Berg Avala to the south of Beograd in Serbia, Neumarkt in the Karavanken, and Montenegro.

894

Handbook

Algeria. Algeria has become the second largest mercury producer in the western hemisphere after Spain, and in 1986 ca. 690 t of mercury was produced. Germany. In 1936 a modern mercury works was built at Landsberg near Obermoschel in the northern Pfalz (Palatinate), which had to be closed in 1942 because working the very low-grade ore, containing only 0.1 % Hg, was uneconomical. Austria. Fahl ore containing 1.8% mercury is mined in small amounts at Schwaz in the Tyrol. Other deposits at Dollach on the Drava are no longer mined. Finland. Outokumpu obtains about 100 t of mercury per year from the processing of zinc concentrates. Former Czechoslovakia. Three important deposits exist: at Kotrbachy, mercury is obtained as a by-product of a fahl ore. Cinnabar is found southwest of Gelnica in a workable ore containing 0.25% mercury. Finally, mercury has been obtained in the region between Mernik and Vranov since the end of the 17th century. This production ceased in 1937. Romania. Romania has mercury ores at Zlatna and Baboja in Transylvania. Turkey. In Turkey about 220 t of mercury is obtained annually from mining operations. At present, because of its low price, mercury is extracted from only two mines, which belong to the state-owned Etibank. The deposits are at Izmir-Odemis-Halikoy (1 X 10 t of ore containing 0.25% mercury), Konya-Ladik, and Konya-Sizma (1.15 x 10 t of ore containing 0.23% mercury). These deposits alone account for ca. 40% of the Turkish reserves of mercury [7]. 6

6

Former Soviet Union. The most important deposits in the former Soviet Union are at Nikotovka in the Ukraine. Per year about 400 t of mercury is produced there from an antimonite ore containing 0.4% mercury. Cinnabar deposits have recently been discovered in the Crimea, although no further details are available. Additional ore deposits are mined in the

of Extractive

Metallurgy

Northern Caucasus, Urals, Altai Mountains, and in Turkestan and Dagestan. China. China has important deposits in the provinces of Yunnan, Hunan, and Kweitschan. The mines at Wanschantschang and Patschai are well known. Mined production was ca. 760 t of mercury in 1987. Japan. In Japan, mercury is extracted on the island of Hokkaido. Canada. Mercury ore deposits are situated in the west of Canada in British Columbia at Pinchi Lake and in the vicinity of Vancouver (Kamloops, Bridge River). United States. The number of mines in the United States has decreased considerably in the last 20 years. Whereas 149 mines were operating in 1965, only 3 mines were still operating by 1981. The principal mining regions are California and Nevada. Placer, Inc., a subsidiary of the Canadian company Placer Development, accounts for ca. 99% of production. The average mercury content in the ore is 0.3— 0.5%. Mining production in 1986 was ca. 400 t and falling. Mexico. The 200 known mercury ore deposits are distributed over 20 states. The most important mines are in the provinces of Zacatecas, Guerero, Durango, Chihuahua, Guanajuato, San Luis Potosi Aguascalientes, and Queretaro; total mercury production is some 500 t/a. South America. South America is of only minor importance as a mercury producer. Chile occupies first place among the mercury-producing countries. The Santa Barbara mine at Huancavelica in Peru is now exhausted, although it supplied substantial amounts of mercury in the 17th and 18th centuries: some 50 000 t was used for precious-metal extraction. Small amounts of mercury are mined in Venezuela and Bolivia. A U.S. estimate made in 1983 gave the following amounts of mercury extracted by mining operations [8]: Spain

50 000 t

China Soviet Union Yugoslavia

17 200 t 17 200 t 17 200 t

Mercury

895

Italy United States Other countries World total

12

100 12 100 29 000 154 800

t t t t

Mining activities and exploitation of deposits are, as with most base metals, greatly dependent on current world market prices. Too low returns have led in recent years to a number of mines being closed, which has resulted in a concentration of plants (less plants with increased production).

17.3.2

Secondary Sources

Like primary ores, industrial waste containing mercury also contributes to its production. The majority of plants used in chlor-alkali electrolysis employ liquid mercury cathodes, resulting in residues containing 10% mercury or more. In addition to this major secondary source, mercury batteries, mercury fluorescent tubes, electrical switches, thermometer breakage, and obsolete rectifiers should be regarded as mercury sources. Although the overall production of mercury has decreased over the last 20 years, sufficient potential uses and thus secondary sources remain for the foreseeable future, thanks to the unique properties of this metal.

17.4

Production

17.4.1 Extraction from Primary Sources

the ores are comminuted to grain sizes of 0.075 mm (200 mesh) in the case of antimony and 0.5 mm (25 mesh) in the case of arsenic. After comminution, lead(II) nitrate [Pb(NO ) ], butyl xanthate, and pine oil are added to the first flotation stage. Flotation is performed at pH 7.1. After purification, mercury is selectively floated, and a concentrate of ca. 50% antimony is obtained. Potassium dichromate (K Cr O ) is added as reactant. The end concentrate has a mercury content of ca. 7 0 % [9]. Another way of concentrating mercury by flotation is to add potassium oleate as collector [10]. 3

2

Beneficiation

Preliminary concentration is desirable, especially for working low-grade ores. This is often performed in the mine by classification. Because of the brittleness of cinnabar, pieces of ore break easily at the mercury sulfide veins. Crystalline mercury sulfide in brittle ore can be concentrated by using settling tables. Another method of concentration is the use of a flotation stage to increase the mercury content. Separation of the antimony and arsenic fractions is particularly important, and

2

7

Only a coarse preliminary grinding to ca. 50-mm grain size is necessary for high-concentration mercury sulfide ores. During subsequent thermal treatment, the ore particles burst because of the high vapor pressure, resulting in further comminution.

17.4.1.2 Processing to Metallic Mercury [11,12] The most important starting material for mercury extraction is mercury sulfide (cinnabar, cinnabarite); it is nonpoisonous and can be stored and transported without any problem. The coarsely ground ore is processed in directly or indirecdy heated furnaces, retorts, or muffles. Reaction with oxygen begins at 300 °C, according to HgS + O

17.4.1.1

2

SO + Hg

2

2

Quicklime can be added as flux to bind the sulfur in solid form. The overall reaction in the absence of oxygen is as follows: 4HgS + 4CaO

3CaS + CaSO + 4Hg 4

In the presence of oxygen, the reaction is 3

HgS + CaO + / O 2

2

Hg + CaSO

4

Addition of metallic iron in the form of iron filings enables the following reaction to occur: HgS + Fe

FeS + Hg

Instead of metallic iron, iron oxides can also be used:

896

Handbook

7HgS + 2Fe O 2

3

4FeS + 3SO + 7Hg 2

For this reason, sulfur-resistant steel should be used as construction material. Processing is normally performed with the addition of air, and the sulfur dioxide formed can be converted in a downstream wet scrubber.

17.4.1.3

Furnace Systems

Mercury sulfide-containing ores are processed pyrometallurgically. Distinction is made between directly and indirectly heated furnaces. This is important because the formation of stupp cannot be avoided in pyrometallurgical processing (the word stupp is an Austrian expression that simply means powder; in the case of mercury, a mercury-containing powder). The greater the formation of dust in the furnace, the greater is the occurrence of stupp. Because of the presence of combustion air in the furnace, directly heated furnaces generally create more dust than indirectly heated ones. Mercury in vapor form condenses to some extent on the small dust particles which act as condensation nuclei. Another difference between directly and indirectly heated furnaces is that mercury vapor in indirectly heated systems is more highly concentrated because it is not diluted by combustion gases. Directly Heated Furnaces. Cylindrical Bustamente furnaces, multiple-hearth furnaces, and rotary kilns are used. The Bustamente furnace is still used in Spain (Almadén) [5]. The cylindrical vessel is heated outside the shaft, only the combustion gases come into contact with the ore. A historical feature is that mercury vapors are condensed in air-cooled terracotta pipes. The multiple-hearth furnace is advantageous for working up rich concentrates, mercury waste of a similar type, or stupp. This kind of furnace has been used for many decades in various mercury mines in North America, and in Almadén most of the ore extracted since 1961 has been processed in multiple-hearth furnaces. The hearths and outer walls of the furnaces are built of shaped fire-

of Extractive

Metallurgy

clay bricks. The furnaces are lined with sheet steel. Cast steels with additions of titanium, chromium, and nickel have proved suitable for internal structural elements (stirrer shaft, raking arm). A modern furnace is up to 10 m high, has a diameter of ca. 5 m, and contains up to 16 hearths that are heated with wood, gas, or oil (mainly the upper and lower hearths). An attempt is made to keep the flow velocities for the reaction gas low (maximally 3 m/s) by skillful feeding of the material (if possible by special sliding surfaces at the throats in the case of small hearth interspacings) and by separately installed gas passageways. The amount of flue dust is thereby lowered (0.2-3% of the feedstock), and the formation of stupp is minimized. Addition of combustion air through the insulated hollow shaft and raking arm and, if necessary, preheating of the fuel oil both reduce fuel consumption. By accurate metering of the inlet air, a lower amount of gas can be passed through the furnace, and the capacity of the subsequent condensation unit can be reduced. Table 17.4 shows some operational examples of multiple-hearth furnaces. With regard to environmental protection, multiple-hearth furnaces are relatively difficult to make gastight and, in some cases, have had to be shut down. Rotary kilns have been used since 1913 to extract mercury. Rotary kilns and rotary drum furnaces provide a high mercury yield with a good throughput rate of the ore because the reaction temperature can be controlled accurately; the ore does not have to be specially preheated; and material throughput can be regulated. Table 17.4: Operational data for multiple-hearth furnaces for mercury extraction. New Pershing Almadén Furnace diameter, m Number of hearths Worked material Throughput, t/d Fuel consumption per tonne of material

Quicksilver Mines

4.9

4.9

6

8

6

ore

ore

ore

5.5

38

77

50

47 L oil

21 L oil

0.15 m wood

3

Mercury

897

The ore is ground to about 65-mm particle size before it is charged. Lime, charcoal, or low-temperature coke may be added as fluxes, and the furnaces are heated directly by oil or gas. The furnaces are operated at 320-400 °C at the charging head and 700-800 °C at the discharge end. For normal furnace sizes and ores, heat consumption is (1 to 1.25) x 10 kJ per tonne of throughput. Without the addition of coal as flux, this corresponds to up to 30 L of oil or 250 m (STP) of fuel gas per tonne of ore throughput. A disadvantage of rotary kilns is the dust formed by the intense material movement, which leads to increased stupp formation in subsequent condensation and can cause stoppage in the waste-gas line. Flue dust amounting to 0.75-6% of the ore feedstock by weight is observed. In many cases, dust formation can be reduced by fluxes having a sintering effect. Some operating examples of older rotary kilns are listed in Table 17.5. 6

aligned shaft. A rotary kiln constructed from two concentric tubes was developed several years later. The ore passes continuously through the inner tube; the outer tube is heated. A highly thermally conducting, gastight silicon carbide is used as cladding. Several proposals for processing mercurycontaining material in indirectly heated furnaces have been made, and patents have been issued [13-15].

3

Indirectly Heated Furnaces. The basic advantage of indirect heating is the lower gas velocity in the reactor and the higher mercury concentration of the process gas. The furnaces are heated by gas, coal, or by electricity and consist in some cases of clay muffles, vessels lined with fireclay or ceramics, or iron vessels lined with silicon carbide. F. Krupp-Grusonwerk AG has arranged several tubular individual muffles in a rotary kiln around a thermally insulated, centrally

The subsequent condensation of mercury from the furnace off-gas is treated later.

17.4.2 Extraction from Secondary Sources Because of the relatively high toxicity of metallic mercury and some of its compounds, spent products must be reprocessed to a large extent. Storage of slurry-like residues is complicated and expensive, and is governed by the limited storage capacity of the closed mines. Active carbon slurries from the effluent concentration resulting from chlor-alkali electrolysis constitute the major proportion of mercury residues that must be reprocessed. In addition to these residues, the processing of fluorescent bulbs, which contain about 1550 mg of metallic mercury per lamp, is becoming increasingly important. Several plants dealing specifically with the disposal of these lamps already exist in Germany [16].

Table 17.5: Operational data for rotary kilns for mercury extraction. Gelnica

Abbadia San New Nevada, Quick- Pershing, Quick- Mercury MinSalvatore Indria silver Mines silver Mines ing Company

Furnace length, m

14

16

17

12.2

18.3

21

Internal diameter, m

1.5

1.25

1.22

0.92

1.22

1.22

Thickness for lining, mm

150 45

50

250

200

Gradient, %

4

5

Time for one revolution, s

72

Moisture content of ore, %

5-10

Hg content of ore, %

0.2

Grain size of ore, mm

50

5

Ore throughput, t/d

40

100

Fuel consumption per tonne of ore

3

0.6-0.8

100 m (STP) gas 100 kg wood + 0.3 m wood 3

4 42

64

40

125

40

45

50 45

19L oil

26.5 L oil

30 L oil

29 L oil

898

Handbook

of Extractive

Metallurgy

from waste gas most efficiently by adsorption on activated carbon or similar substances, mercury-containing charcoal from waste gas purification will also require reprocessing in the future.

Figure 17.1: Reduction of mercury in primary batteries in Germany [17]: a) Mercury in sold primary batteries; b) Mercury in domestic refuse.

The mercury content of normal zinc-manganese dioxide batteries has been reduced considerably over the last few years (addition of mercury inhibits the formation of gas by zinc). In Germany, the mercury content of batteries will be reduced to 0 . 1 5 % by 1 9 9 0 (Figure 1 7 . 1 ) , starting from a level of 1 % [ 1 7 ] . In addition to zinc-manganese dioxide batteries, zinc-mercury(II) oxide batteries are also widely used as round cells. These contain ca. 3 0 % mercury, which must be recovered. Mercury in concentrations of ca. 1% is also found in zinc-silver(II) oxide and zinc-oxygen batteries. A particular problem in reprocessing batteries is grading and sorting the various types, which are inadequately identified and, in some cases, have the same external dimensions for different electrochemical systems. Treatment processes have been developed in Japan [ 1 8 ] , Germany [ 1 9 , 2 0 ] , and other countries [21], which can remove mercury by treatment in rotary kilns or in a distillation chamber. The individual types of batteries do not have to be sorted beforehand, and a mixture of batteries can be processed. Because batteries also include a large number of plastics (e.g., as sealants), waste gases must be treated and burned if necessary. Because a proportion of spent batteries end up in domestic refuse, high mercury concentrations can be found in waste gas from domestic refuse combustion. Mercurycontaining older printing inks are another source. Because mercury can be precipitated

The mercury-selenium residue formed in the primary smelting of mercury-containing sulfidic ores requires special reprocessing. This substance, which occurs as a slurry, has been reprocessed in a multiple-hearth furnace in which the mercury fraction is evaporated [ 2 2 ] . Another company converts the residue to metallic mercury in a rotary kiln by adding lime fluxes, with a relatively inert residue remaining behind [ 2 3 ] . A hydrometallurgical reprocessing treatment has been investigated on a pilot scale in a research project [ 2 4 ] . Mercury is extracted in the form of mercury(II) oxide and sulfide, and can be added to the conventional extraction processes. The method does not appear to be economically practical at present because of the relatively expensive reactants involved. Additional secondary sources, such as thermometer breakage, electric switches, and amalgams, can generally be worked up by conventional distillation methods [ 2 5 , 26]. A comprehensive monograph on the topic of evaporation and thermal dissociation of mercury sulfide is available [ 2 7 ] . Electrolysis has been proposed for the removal of mercury from gold-containing solutions: 9 0 - 9 5 % mercury and < 1 0 % gold are deposited at 1 . 0 - 1 . 5 V [28].

17.4.3 Condensation of Mercury from Furnace Off-Gas The vapor pressure p (mbar of mercury) can be calculated according to a relationship given by BARIN and KNACKE [ 2 9 ] :

where T is measured in kelvin. The degree of saturation s (kg of mercury per cubic meter) of the gas at t ( ° C ) or T ( K ) and p = 1 0 1 . 3 kPa can be determined from the

Mercury

899

pressure and density of mercury vapor (3.9091 3 kg/m at 357 °C) (α = 1/273):

6

require a heat input of ca. 1.2 x 10 kJ per tonne of ore. The following amounts of waste 3 gas [in m (STP)] are produced per 1000 kJ for the different fuels with 20% excess of air: Wood

The degree of saturation s of mercury vapor calculated according to this formula for differ­ ent temperatures is summarized in Table 17.6. Figure 17.2 shows p and s plotted as a function of 1/T on a semilogarithmic scale. Mercury losses during condensation can be calculated as follows: 1 m3 of saturated vapor at 140 °C (413 K) enters the condensation unit laden with 14.464 g of mercury. There it is cooled to 20 °C (293 K) (s = 0.0141 g/m3), and a volume contraction to 293/413 = 0.71 m 3 oc­ curs. The saturated vapor leaving the unit therefore contains 0.71 x 0.0141 = 0.01 g of mercury. The loss of uncondensed mercury re­ moved with the waste gas is thus 0.07% of the initial amount. With a condensation end tem­ perature of 40 °C and otherwise identical con­ ditions, the loss increases to 0.35 % Hg. Table 17.6: Saturation content of mercury vapor at various temperatures. Saturation content, s

p, kPa

T, K

g/m

g/m (STP)

7.03 x 10-5 1.71 x 10-4 3.919 x 10-4 8.514 x 10-4 1.761 x 10-3 3.486 x 10-3 6.625 x 10-3 0.012138 0.02150 0.03690 0.06154 0.100 0.1585 0.2457 0.3727 2.2894 3.8743 32.808

283 293 303 313 323 333 343 353 363 373 383 393 403 413 423 473 523 573

0.00604 0.01406 0.03144 0.06612 0.1325 0.2545 0.4695 0.8359 1.440 2.404 3.906 6.187 9.558 14.464 21.417 117.65 458.97 1391.5

0.00626 0.01508 0.03489 0.07588 0.1568 0.3104 0.5893 1.081 1.915 3.284 5.478 8.906 14.110 21.385 33.175 203.65 879.25 2921.0

101.3

630

3909.1

9021.1

3

3

In practical operation the mercury content in the hot reaction waste gases is always far below the saturation limit because the theoret­ ical vapor-liquid equilibrium is not reached. Directly heated mechanical roasting furnaces

0.387

Generator gas

0.42

Oil

0.32

Per tonne of ore, this corresponds to about 470 m (STP) with wood or generator gas firing and 375 m3 (STP) with oil firing. With 5 % wa­ ter content in the ore, 70 m 3 (STP) of steam is produced in addition. If an ore containing 0.3% Hg is worked up, the mercury content with wood or gas firing is 5.56 g/m 3 (STP), and with oil firing 6.75 g/m3 (STP), at 140 °C. With a condensation end temperature of 20 °C at which the gas still contains 0.0141 g of mercury per cubic meter corresponding to the saturation limit a loss of 0.21-0.25% mer­ cury occurs, and at a condensation end tem­ perature of 40 °C, a loss of 1.1-1.3 % mercury from the waste gas occurs. In general, the mercury residual content in waste gas, compared to the amount at 20 °C, increases with a temperature increase of 10 °C by a factor of 2.2 20 °C by a factor of 4.7

Figure 17.2: Vapor pressure p and saturation content s (g/m3) as a function of temperature; s s saturation content under standard conditions [g/m3 (STP)].

900

Handbook

30 °C by a factor of 9.4 40 °C by a factor of 18.1 50 °C by a factor of 33.4

Thus efficient cooling of the waste gas is necessary. In a plant that processes an ore with 5% moisture content, as much as 5 0 % of the cooler capacity is used for condensation of evaporated moisture alone, without the steam originating from fuel combustion being taken into account. Several practical requirements must be met for satisfactory operation: (1) The ore should be as dry as possible. (2) The amount of waste gas produced by the fuel should be low. (3) When the gas enters the condenser, its temperature should be only 1 0 - 2 0 °C above the mercury dew point. (4) The outlet temperature should be as low as possible. (5) Before entering the stack, the reaction gas should pass through an afterpurification section, if possible. Stupp formation can be kept low by (1) indirect heating, (2) separate processing of finegrain classes of ore, (3) predrying of ore that is too wet, or (4) wetting of ore that is too dry. The flue dust should be separated as completely as possible from the still hot reaction gas before it enters the mercury condensation unit. A conveniendy arranged condensation system for a directly heated mercury smelting plant operating with mechanical roasting furnaces consists of the following parts: • Dust separator in the form of a cyclone, an electrostatic precipitator, or a combination of both • Fan (e.g., of Monel Metal) • Cooling unit for mercury condensation • Afterpurification chamber for mercury residues

separating

• Injection chamber for separating traces of mercury • Fan (e.g., of Monel Metal) The cooling unit is itself in many cases also constructed of monel metal, and consists of tubular condensers (CERMAK type) formed from six to ten trains arranged in parallel, each with eight inverted U-tubes of elliptical cross section whose arms terminate in acutely taper-

of Extractive

Metallurgy

ing, water-filled deflection boxes. The tubes are sprinkled with water. The mutually corresponding tubes of all the individual trains in each case end in a common box, and in this way the stupp can be classified. Gate valves in each train enable the draft to be regulated or even permit the train to be separately disengaged. The ratio of tube cross section to fan output must be matched so that the gas flow velocity is maximally 0.75 m/s in narrow parts of the tubular condenser. The overall unit must be designed so that half of the tube trains can be decommissioned for cleaning and repair, whereas the remaining tube trains ensure satisfactory cooling of the reaction gas while maintaining draft conditions in the furnace. Gases from indirectly heated furnaces that contain only a small amount of flue dust can be condensed by spraying water directly into the mercury-containing waste-gas stream. To ensure sufficient mercury condensation, the gas must pass through a succession of water curtains. Fewer parallel tubes of larger diameter are used in the cooling unit for directly heated furnaces. The last stage operates with fresh water; for the other cooling stages, water is circulated and recycled through suitably dimensioned settling tanks. All setting tanks within the condensation unit have sloping floors and connecting pipes through which mercury metal flows to the lowest part of the system. The tanks stand on feet on a smooth cement floor, which slopes so that any leakage can be detected and escaping mercury collected. The material for these tanks is wood or cast iron; stupp-collecting tanks are also constructed of concrete. When the plant is operating continuously, all places prone to thick deposits must be mechanically cleaned once a week; the remaining areas can be cleaned at longer intervals.

17.4.4

Treatment of the Stupp

The amount of mercury contained in the stupp may represent a substantial proportion of production. In Idrija, when processing highbituminous ore, for example, 7 8 - 9 1 % of the

Mercury

mercury is extracted from the stupp and only 9-22% is extracted directly. To extract mer­ cury from the stupp, individual mercury drop­ lets must be coagulated, which can generally be achieved by kneading and pressing the mass. Stirring the stupp with sievelike perfo­ rated rakes in iron vessels also gives good re­ sults. Addition of quicklime to the stupp neutralizes any acids contained in the conden­ sate, saponifies any fats present, absorbs a large proportion of the water, and chemically reduces mercury sulfate. Mercury and dust can also be mechanically separated by simple wash treatment with wa­ ter. A tenfold mercury enrichment from the stupp by flotation was described in 1929 [30]. This very effective method of separating mer­ cury enables the waste material to be dumped directly onto the waste tip in certain cases. Generally, however, extraction of mercury from the stupp is not so effective, with the re­ sult that the low-mercury fraction must be re­ turned to the smelting process. For regularly occurring large amounts of stupp, a special furnace unit is worthwhile (retorts or multiplehearth furnace).

17.5 Environmental Protection 17.5.1 Natural Distribution of Mercury Because of its high vapor pressure, metallic mercury disperses relatively quickly into the atmosphere and, with suitable air movement, is taken up by plants and animals. The average concentration in the earth's crust is 0.08 ppm and in seawater 3 x 10-5 ppm. Mercury is thus one of earth's relatively rare elements. The natural mercury content of the atmosphere is 0.005-0.06 ng/m3 [31]; in plants, 0.001-0.3 μg/g (generally < 0.01 μg/g); and in meat, 0.001-0.05 μg/g [32]. A comprehensive list of mercury in the environment appears in [33]. A quantitative summary of the occurrence, distribution, and utilization of mercury is given in the following material [34]:

901

Mercury content of the oceans 70 000 000 t Mercury content of the earth's crust (1-m-thick layer) 100 000 000 t Natural atmospheric mercury emission (volcanoes, wind erosion, degassing) 25-100 000 t/a Natural mercury emission in water (weathering in rivers) 5000 t/a Use of fossil fuels (minerals, deposits) 8-10 000 t/a Use of mercury and its compounds 6000 t/a

Natural air emissions from mechanical ac­ tivity, wind erosion, and degassing constitute the largest proportion of emitted mercury. By contrast, the utilization of mercury is rela­ tively small, although this should not mini­ mize or obscure the danger of mercury at high concentration.

17.5.1.1 Mercury in Soil, Plants, and Animals Relatively large amounts of mercury are circulated due to the constant exchange of mercury among water, soil, and the atmo­ sphere. Some of this migrating mercury is re­ tained in the soil in the form of humus compounds and, in certain cases, is also con­ centrated [35]. The mercury content of various soils in Austria unless otherwise stated (mg per kilogram of soil) is as follows [36, 37]: Meadowland in Arable land in

Niederösterreich Niederösterreich

0.039 0.180

Vineyards in the

Burgenland

0.080

Reed belt on the

Neusiedlersee

0.340

Arable land (3 % humus) in Kremsmünster 0.065 Meadowland (4.8% humus) in Kremsmünster Woodland (10.2% humus) in Kremsmünster Garden soil in England Rice field slurry in Japan Fields for agricultural use in Niedersachsen, Germany Woodland in Niedersachsen, Germany Mean mercury value for Europe (10-cm-thick humus layer)

0.070 0.170 0.25-15.0 0.40-1.8 0.055-0.104 0.249-1.672 120 g/ha

Organically bound mercury is also added to the soil with seeds as a seed treatment agent (fungicide). Amounts of mercury used for seed treatment or soil disinfection are as follows: Grain (wheat) ca. 7 g/ha added as seed treatment agent to the soil per 150 kg of seed material/ha Sugar beet (seed material in pill form) ca. 1 g/ha added as seed treatment agent to the soil per 200 000 pills/ha Cotton (seed treatment and disinfection) ca. 20 g/ha

902

Handbook of

Extractive

Metallurgy

regardless of whether the soil has been anthropogenically contaminated [46]. With more serious contamination, especially atmospheric, the mercury content of fungi and mushrooms increases considerably [47].

Figure 17.3: Microcirculation of mercury (in g h a - l a - l ) in the soil [36]. Lawns (disinfection)

ca.

80 g/ha

However, neither soils that have received mercury-treated seeds for many years, nor plants grown on such soils, exhibit a high mercury content [35]. As pot tests have shown, the mercury is partly retained in the humus and partly evaporates from the soil [38, 39]. The amount of residue passing into groundwater is insignificant [40] (Figure 17.3). According to investigations carried out in Norway, the mercury content of soil varies between 0.022 and 0.55 ppm (mean value 0.19), without any "recognizable" effect due to human activity [41]. The mercury content in soil is higher in geothermal or ore-containing regions and can be as high as several parts per million. Thus, mercury is often used as a tracer metal in geological exploration [42]. Mercury content has been investigated in soil and lichen in the mercury mining region around Monte Amiata (Italy). The mercury content in the soil and in the lichen decreases sharply with increasing distance from the mine. Mine waste gases and worked ore are the main sources of emission [43]. Mercury in Fungi and Mushrooms. The fleshy parts of fungi and mushrooms can contain unexpectedly high concentrations of mercury [44, 45]. Because of the affinity of mercury for the sulfur-containing constituents of fungi, the latter can absorb and concentrate mercury from the local soil. Mercury-rich fungi and mushrooms are found everywhere,

Mercury in Fish and Seals. The mercury content in fish varies depending on type and habitat [48]. The assumption that fish concentrate mercury to 2000 times its concentration in the ambient medium is neither valid nor statistically relevant and has no significance as a rule of thumb [48]. Mercury uptake in warm, tropical waters is higher than in cold northern waters [49]. Because of the concentration of mercury in sediment, dabbling fish, which collect their food from the bottom of the water, absorb more mercury than nondabbling fish. Predatory fish (e.g., pike) contain higher than normal levels of mercury only if they preferentially feed on dabbling fish. The controversy surrounding canned tuna in the United States has revealed that tuna preparations in the Smithonian Institute from ca. 1880 have the same mercury concentration as present-day fish. Under the assumption that the values found in museum preparations constitute a lower limit—some of the mercury might well have evaporated over the course of time—the fact that present-day values are the same as previous values means that no concentration has occurred. Data from the Danube upstream of large cities and industrial regions, where trout have been found with a mercury content of > 1 mg of mercury per kilogram of fish, point to factors other than industrialization. The high mercury concentration in the liver of seals in the northern section of the east coast of North America (Labrador), where no large cities or industrial regions are to be found, is attributed to underwater volcanic emissions.

17.5.1.2

Mercury in Food

Many investigations were carried out on foods from the 1920s and 1930s onward [50]. In Germany, the Zentrale Erfassungs- und Bewertungsstelle für Umweltchemikalien (central office for collecting and evaluating data on

Mercury

903

environmental chemicals; ZEBS) of the Federal Ministry of Health in Berlin in particular, has collected data from 1975, 1979, and 1982. According to these data, if average consumption patterns are assumed, mercury uptake in 1979 was about 0.052 mg per week for the "average" inhabitant of Germany (i.e., only about one-sixth of the upper limit recommended by WHO). The major proportion of mercury is absorbed from animal foods [51]. The legal limits are given in Table 17.7. The most recent ZEBS investigations [52], covering 1978-1982, are presented in Tables 17.8 and 17.9. The 1982 values essentially confirm the analytical data of 1979. The values for milk, condensed milk, cheese, and eggs; veal, beef, pork, calves liver, calves kidneys, poultry, sausages, and meat products; vegetable oils and fats; rice, rye, and potatoes;

leafy, sprouting, fruit, or root vegetables and canned vegetables; pomaceous, stone, berry, citrus, shell, or canned fruit; fruit and vegetable juices; and beverages such as wine, beer, and chocolate are all < 0.020 mg of mercury per kilogram of fresh substance and, in most cases, < 0.010 mg/kg. The average amounts of mercury actually absorbed per week are compared in Table 17.9 with the limiting values specified by WHO. Foods contain the major proportion as follows: drinking water (14.8%), wheat (12.9%), milk (10.2%), mineral water (8.1%), canned fish (8.0%), coffee (7.2%), potatoes (6.0%), and fish (5.6%). Another detailed investigation carried out by ZEBS is concerned with the mercury, magnesium, and zinc content of breast milk, blood serum, and the fatty tissue of nursing mothers [53].

Table 17.7: Legal limiting values for total uptake of mercury (1980). Fish, mg/kg

Total uptake, mg/week

Switzerland

0.5

Japan

0.4 (Hg) 0.3 (Hg methyl)

0.17 mg of Hg per 50 kg of body weight

United States

1.0

0.005 mg/kg ofbody weight (Hg) 0.0033 mg/kg ofbody weight (Hg methyl)

Sweden

no limit (lakes that contain fish with more than 1 ppm Hg are placed on ablacklist); fish with a high Hg content may be legally rejected.

Germany

1.0

WHO recommendation

0.3 (max.)

Table 17.8: Mercury in food. Content, kg/kg of fresh substance

Food

1982

1979

Cow's liver

0.021

0.015

Pig's liver

0.047

0.058

Cow's kidneys

0.077

0.066

Pig's kidneys

0.246

0.260

Freshwater fish

0.271

0.257

0.196

0.127

1.070

0.859

Saltwater fish, except Hg-susceptible fish Hg-susceptible fish

a

Fish products

0.208

Canned fish

0.206

0.188

Wheat

0.026

0.003

Coffee

0.041

Table 17.9: Uptake amounts of mercury in absolute figures and as a proportion of W H O values. Hg

Hg-susceptible fish include: mackerel shark, dogfish, blue ling, halibut, black halibut, turbot, and Greenland shark.

0.35 (0.290)

Men (70kg) mg/week Proportion o f W H O figure, %

0.1229 35.11

Women (58 kg) mg/week Proportion of W G O figure, %

0.0933 32.17

a

a

a

a

W H O value mg/week (70 kg) mg/week (58 kg)

The median values of toxic substance levels were taken into account.

904

17.5.2 17.5.2.1

Handbook

Mercury Emissions Gas Purification

of Extractive

In the second method, the cooled and dusted roast gas is treated with a mercury(II) chloride solution, and mercury precipitates as mercury(I) chloride (Hg Cl ). Mercury is removed from the wash liquid by treatment with sodium sulfide. Part of the mercury(I) chloride produced is oxidized to mercury(II) chloride with gaseous chlorine and returned to the process. The final level of mercury in the waste gas is 0.05-0.1 mg/m [55, 56]. 2

In a large number of combustion processes, mercury is transferred to the gaseous phase and must be removed again by subsequent treatment. The largest amounts of secondary mercury occur in the smelting of sulfidic ores. Mercury which is present as mercury sulfide is released during roasting. At present, two methods are primarily used for waste-gas purification.

2

3

In a third method, waste gas containing mercury and sulfur dioxide is treated with a wash solution containing Cu and Hg ions in addition to H S 0 (200-300 g/L) and HC1 (5 g/L). Mercury in the vapor is thereby oxidized to Hg , and the Hg values of the waste gas are 0.02-0.05 mg/m [57]. 2+

In the first method, the gas is treated with sulfuric acid (90%) at 200 °C. The mercury(I) sulfate formed is deposited in wash towers. Fine purification is performed by afterpurification with sodium sulfide [54].

Metallurgy

2

2+

4

+

0

3

Figure 17.4: Specific mercury load distribution in the Bamberg refuse heating and power station with TMT-15 precipitation, with and without sludge combustion [63].

Mercury

905

Another source of mercury emission is refuse combustion. Mercury occurs in the form of thermometer breakage, fluorescent tubes, switches, and batteries in domestic refuse. Investigations have shown that mer­ cury in flue gases is present as mercury(II) chloride [58]. Below ca. 150 °C, the flue ash contained in flue gases adsorbs mercury(II) chloride and reduces it to mercury(I) chloride. Active carbon or active carbon impregnated with iodine compounds can also be used as a mercury adsorbent [59-62]. The mercury con­ tent of domestic refuse is 3-5 g/t [63]. The percentage distribution of mercury in a typical refuse combustion plant is shown schemati­ cally in Figure 17.4. Most of the mercury passes into the residue to be dumped.

17.5.2.2

Water Purification

The solubility of mercury in water depends strongly on the temperature. It decreases with decreasing temperature and can be reduced further by addition of salts (e.g., sodium chlo­ ride). A whole range of equipment and devices can be used in purification methods to remove mercury from water. Examples are activated carbon filters, ion exchangers, and electrolysis systems. Mercury can also be removed from solution by addition of suitable reagents. The level of 10-30 mg of mercury per liter in wastewater from chlor-alkali electrolysis can be reduced to 0.1 mg/L by using activated carbon and graphite powder. The carbon has a grain size of 5-100 μm [64, 65]. In principle, all types of activated carbon can be used for the fine purification of mercury-containing water. Carbon is dispersed in the water and then filtered. For ion exchangers, a distinction can be made between reusable and disposable ones [65-67]. In some cases, sulfur-containing sub­ stances are used as active groups, which re­ move mercury from the solution in the form of a mercury sulfide compound. Copper fluidized-bed electrolysis provides another method for mercury separation. The copper powder cathode is held in suspension

by the flowing solution, and mercury is depos­ ited as an amalgam. The copper amalgam is purified by distillation [68]. By adding small amounts of oil which are dispersed in the aqueous solution, mercury can be concentrated in the oil phase. The mer­ cury content of the purified aqueous solution is in some cases < 10 μg/L. The oil is sepa­ rated by centrifugation, and line purification is carried out by using a conventional oil-water emulsion splitting unit. The initial mercury concentration should be between 0.5 and 2.0 mg/L [4]. Tin(II) chloride dissolved in dilute hydrochloric acid has been proposed for treat­ ing mercury ion-containing wash water from combustion units. The reducing agent tin(II) chloride is added in above-stoichiometric quantities. Reduced mercury is added together with the wash water to an evaporation device and is expelled by passage of stripping gas at elevated temperature. Condensation is ef­ fected by cooling the gas stream. The mercury content is thus reduced from 5 mg/L to < 0.1 mg/L, depending on the concentration of tin(II) chloride [69]. To purify flue gas wash water from refuse combined heating and power stations, the ad­ ditive TMT-15 developed by Degussa is used in some cases on an industrial scale (Figure 17.4). The active substance of the additive consists of trimercapto-S-triazine in the form of a sodium salt. Mercury is bound as C 6 N 6 S 6 Hg 3 , The compound is stable up to 210 °C and is only sparingly soluble in the elution test (a test to determine the soluble components of a solid material). The mercury content is reduced from ca. 4 mg/L to < 0.05 mg/L [70, 71]. The additional costs involved in TMT-15 precipitation amount to ca. 0.25 DM per tonne of refuse. Mercury can be removed from concen­ trated sodium hydroxide solution by the use of ultrasound when the solution is filtered. The initial mercury content of ca. 20 mg/kg of aqueous sodium hydroxide is reduced to ca. 0.4 mg/kg after filtration [72]. The fine purification of crude phosphoric acid can be performed at pH 0.5-1.5 by using a diorganodithiophosphorus compound in

Handbook

906

conjunction with activated carbon as adsor­ bent. The final mercury concentration is < 0.02 μg/g of solution [73]. Atpresent, no stan­ dard methods exist for removing mercury from sulfuric acid [74].

17.6

Quality Specifications

Fine Purification of Metallic Mercury [75]. Most of the mercury on the market is 4N mate­ rial (99.99% mercury). Higher purity is sel­ dom required. Impurities in the form of gold ppm) are manifested as dark particles after dissolution in nitric acid.Only a few purifi­ cation methods exist. • Dry Oxidation. In this method, readily oxidizable constituents such as magnesium, zinc, copper, aluminum, calcium, silicon, and sodium can be removed by passing air or oxygen through the liquid metal. The ox­ ides formed have a lower density than mer­ cury and float on its surface. They can be removed by filtration, scooping, or with­ drawing the mercury through an opening in the bottom. • Wet Oxidation. In an aqueous medium, mer­ cury is dissolved by adding nitric, hydro­ chloric, or sulfuric acid with dichromate, permanganate, or peroxide, to oxidize impu­ rities. Good dispersion of the mercury is ex­ tremely important in this method. The aqueous solution can be separated from the mercury by decanting, and traces of water can be removed with calcium oxide. A plant for the wet purification of mercury has been described [76]. • Electrolytic Refining. Perchloric acid con­ taining mercury oxide serves as the electro­ lyte. • Distillation. Mercury can be evaporated un­ der atmospheric pressure or in vacuo. Distil­ lation can be carried out in normal steel vessels or in a glass apparatus. Elements with a lower vapor pressure than mercury can be separated in this way. In many cases, mercury must be distilled repeatedly to achieve the desired purity, particularly if it is

of Extractive

Metallurgy

to be used to produce cadmium-mercury telluride. Additional methods, adapted to the relevant processes, are available for purifying and working up larger amounts of mercury [77, 78].

17.7 Chemical Analysis The oldest and simplest method for deter­ mining mercury in minerals is described by E S C H K A ; it involves a gravimetric method in which mercury is precipitated as an amalgam [79]. The ore sample is weighed in a porcelain crucible, intimately mixed with iron filings, and then covered with a layer of zinc oxide, magnesium oxide, or calcium oxide. The cru­ cible is closed with a tightly fitting cover of gold foil having a cup-shaped depression in the middle, which is sprayed from above with cooling water. On careful heating of the sam­ ple, mercury is distilled and deposits on the underside of the cover. When a constant weight is reached at the gold foil, all the mer­ cury has been collected. Measurements can be affected by cadmium and arsenic, which also condense on the gold foil. This method of mercury determination requires a high level of experimental skill and care on the part of the analyst, whose technique greatly affects the accuracy and reproducibility of the results. The reliably detectable minimum mercury content is between 10 and 50 ppm. Theoreti­ cally, the accuracy could be improved by in­ creasing the amount of sample, although this would give rise to difficulties with regard to the apparatus. A method for precipitating monovalent mercury as an iodate from a neutral or weakly acidic solution (nitric acid) is described in [80]. Mercury can be determined gravimetrically after the mercury(I) iodate precipitate is washed with ethanol and diethyl ether. Parallel gravimetric determination of the excess of po­ tassium iodate precipitation agent is possible with thiosulfate after the mother liquor has been acidified with sulfuric acid and potas-

Mercury

sium iodide has been added. Both methods give good results. A quick method for determining mercury(II) is based on the fact that a complex is formed when an excess of potassium iodide is added to a neutral or weakly ammoniacal mer­ cury solution. This complexes with copper diethylenediamine sulfate to form violet crystals of the complex double salt copper diethylenediamine mercury iodide, which is practically insoluble in ethyl alcohol or diethyl ether [81]. In a rapid classical qualitative method for detecting mercury, ca. 1 g of the substance to be tested is digested with acid; the resultant solution is oxidized with a drop of bromine (the solution must not turn yellow) and then boiled with a few milliliters of the reagent (10 g of KI and 100 g of NaOH in 100 mL of H 2 0 ) and filtered. In the filtrate, mercury is deter­ mined by the black precipitate formed on dropwise addition of a Sn2+ solution [82]. Modern operational monitoring employs physical analytical methods and test tube methods suitable for quick detection. Both Xray fluorescence spectroscopy and atomic ab­ sorption spectroscopy (AAS) have proved suitable for quantitative and qualitative mer­ cury determination. The detection limit for these methods is so low that the maximum workplace concentration values can be moni­ tored precisely [83]. Portable atomic absorption spectrometers, for example, are available, which indicate the atmospheric mercury concentration as an ana­ log or digital display after a short warm-up time of the spectrometer. The result of the analysis is available immediately and simpli­ fies the monitoring of a production plant. The measurement value can also be recorded con­ tinuously at stationary measurement sites with a recorder. The influence of interfering factors in AAS determination is discussed in [84]. Differential pulse anodic stripping voltammetry in conjunction with a gold electrode can be used to detect copper and mercury in natu­ ral water and wine [85, 86]. A mercury con­ centration of 0.02 μg/L can be measured.

907

Mercury in the air can be detected down to 3 0.1 ng/m with the Coleman mercury analyzer system (based on a very selective cold vapor atomic absorption), to an accuracy of 10%. One analysis takes about 3 min [83].

17.8 Storage and Transportation A classification for the transportation of mercury, mercury oxide, mercury(I) chloride, and mercury(II) chloride is given in [87] (specification sheets 831, 863, 865, 868). DORIAS presents a detailed description of the properties, handling, storage, and transporta­ tion of dangerous substances [88], as well as a list of addresses of the relevant authorities (throughout Europe) for information pur­ poses. The provisions differ from country to country and must be ascertained from the rele­ vant authorities. In general, containers of stainless steel, normal-quality steel, iron, glass, ceramics, and a range of plastics are suitable for storing mer­ cury. When storing liquids that contain ex­ tremely low levels of mercury (μg/g-ng/g range) in plastic containers, mercury losses occur with a large number of plastics. When water containing 5 ng/g of mercury was stored in polyethylene bottles, only 5 % of the mer­ cury was present after 21 d. Approximately 77 % of the mercury had been adsorbed on the 2+ side walls, and 18 % had evaporated. The Hg ions are assumed to be reduced to Hg + ions, which in turn disproportionate to Hg0 and 2+ 3+ Hg [89]. Addition of Au ions in trace amounts is sometimes recommended [90]. Further investigations on the storage of mer­ cury in dilute solution are discussed in [9195].

17.9

Uses

Because of its special properties, mercury has had a number of uses for a long time. The conventional application is the thermometer. Mercury is frequently used in pressure gauges and for thermal content measurements.

908

Handbook of Extractive

Metallurgy

Table 17.10: Uses of mercury and its compounds [96]. Area of use

Form

Approximate amount, t/a

Emissions

Other sectors, waste elimination

Remarks

Chemicals, reagents compound

40

Alkali-manganese batteries Mercury oxide batteries

chloride/metal

37

refuse combustion reduced emission by separate collection possible

oxide

20

refuse combustion alternatives still too expensive; separate collection declining

Pesticides

27

Medicinal sector Dental amalgams metal Disinfection compound

26

Fungicides for paints

compound

10

Catalysts

compound

8

Thermometers, barometers, manometers

metal

8

Electrical engineer- metal ing components

6

interval switches Fluorescent tubes

metal

3

Pigments

sulfide

1

laboratories

largely impossible to find suitable replacement; environmental damage can be reduced by other measures

distribution

now only for seed treatment; substitutes are available refuse combustion separate collection possible

weathering

exploitation; removal of old paint coats

declining; in some cases restrictions on use

chemical industry declining instrument and ap- refuse combustion replacement by other methods and processes paratus breakage possible (e.g., electronics) refuse combustion declining; substitution possible largely through electronic components exploitation of scrap refuse combustion demand for fluorescent lamp breakage tubes increasing, despite reduction in mercury use per tube; total use increasing

In mechanical engineering, mercury is used in mercury vapor diffusion pumps for producing a high vacuum. One important area of use is the lighting industry, where mercury is added to various types of bulbs. Among electrical components, mercury switches, rectifiers, oscillators, and primary batteries can be mentioned, which contain up to 30% mercury. Mercury is used as a liquid cathode in the production of chlorine and sodium hydroxide by chlor-alkali electrolysis. Because mercury forms alloys with a large number of metals, mercury alloys have a wide range of applications. Table 17.10 summarizes the areas of use of mercury and its compounds.

refuse combustion declining

17.10

Mercury Alloys

The alloys of mercury (amalgams) occupy a special position among metal alloys because they can be solid, plastic, or liquid at room temperature. Liquid amalgams are true solutions of the alloying elements in mercury, whereas plastic amalgams are suspensions of solid particles of the alloying partners in mercury or a saturated mercury solution. Solid amalgams are intermediary phases, often mixed with alloying partners or their primary mixed crystals. Solid amalgams may contain liquid-phase inclusions. An amalgam is sometimes difficult to identify experimentally. JANGG has suggested a suitable apparatus for synthesizing and analyzing amalgams [97].

Mercury

909

The solubility of many metals in mercury depends strongly on temperature (Figure 17.5). The solubility of some metals in mercury at room temperature is given in Table 17.11. Amalgam formation may be exothermic (e.g., sodium) or endothermic (e.g., gold).

Alkali amalgam is an important intermediate in chlor alkali electrolysis by the amalgam method. Amalgams also play a role in the extraction of cadmium or aluminum [99]. A comprehensive list of thermodynamic data on amalgam formation has been collected by GUMINSKI [100]. After surveying the literature, he is of the opinion that liquid mercury greatly influences intermetallic compound formation. The reactions are comparable to the solid-state formation of ionic and nonionic substances in liquids. Production of Amalgams. Three methods are used for the industrial production of amalgams:

Figure 17.5: Solubility of some metals in mercury as a function of temperature [99]. Table 17.11: Solubility of some metals in mercury at 20 °C [97]. Solubility, %

Metal

42.5 5.0 1.99 147 ca. 0.9 0.62 ca. 0.4

Tl Cd Zn Pb Sn Na K

Metal Mg Au Ag Cu Al Fe Si

Solubility, % ca. 0.3 0.131 0.035 ca. 0.002 0.002 ca. 10 virtually insoluble -5

Ammonium amalgam, which has been known for a long time, is an interesting case of the NH group acting as a metal-like alloying constituent. The synthesis of tetramethylammonium amalgam is described in [98]. Technically important amalgams are those of tin-copper precious metals used in conservative dental treatment. Gold, silver, or tin amalgams are still used for much gold- and silver-plating work, as well as for production of certain types of mirrors. The measurement range of mercury thermometers can be extended to - 5 8 °C by addition of thallium to mercury. 4

• Powder Metallurgy Method. The powdered alloying components are mixed with mercury. The reaction rate of spontaneously occurring amalgam formation is determined by the degree of dispersion of the powder, the rate at which reactants diffuse into one another, and the wettability of the powder by liquid mercury. If other components besides mercury are added simultaneously, they are preferably pulverized as master alloy in the desired weight ratio. In many cases, the powder reacts with mercury more quickly in the presence of a salt of the element to be amalgamated. • Galvanic Method. Many metals can be deposited from their aqueous solutions or from salt melts on mercury cathodes, with simultaneous formation of an amalgam. Because of the high overvoltage of hydrogen on mercury or amalgam cathodes, even nonprecious metals (Zn, Fe, Mn) can be deposited from acid solutions. Nonaqueous solutions may also be used (with Mg, Ti). • Reaction with Sodium Amalgam. Metal exchange can occur in a fast and stoichiometric reaction through the phase boundary by reaction of sodium amalgam with the salt solution of a precious metal (or, generally, by reacting a nonprecious amalgam with a noble-metal salt solution).

910

Handbook

17.11 Compounds Mercury occupies a special position in group 12 of the periodic table. In contrast to the two other members of this group (zinc and cadmium), it forms compounds in two valence states. Compounds of monovalent mercury con­ tain ions in the unusual form Hg 2 + . These com­ pounds are not very stable and disproportionate easily to form elemental mer­ cury and the corresponding divalent mercury derivative. Most of the monovalent com­ pounds are sparingly soluble in water. The more soluble salts, e.g., the nitrate, are par­ tially hydrolyzed in aqueous solution: after acidification of these solutions, the poorly sol­ uble compounds can be obtained by precipita­ tion. In addition, compounds of monovalent mercury can be prepared from those of the di­ valent element by reduction with metallic mercury. The compounds of divalent mercury can be divided into those that are strongly dissociated and those that are weakly dissociated. The former, such as the sulfate and the nitrate, un­ dergo considerable hydrolytic cleavage in wa­ ter. The weakly dissociated compounds, e.g., the chloride and the cyanide, are less prone to hydrolysis by water. With excess anions they form complexes that are more soluble than the salts themselves. The starting material for all of these com­ pounds is elemental mercury: the metal is treated initially with a suitable oxidizing agent, e.g., chlorine or nitric acid. The other compounds can be obtained from the resulting oxidation product by further reaction. Multistep processes are often necessary.

17.11.1

Mercury Chalconides

Chalconides of monovalent mercury are unknown: precipitation from a solution of mercury(I) nitrate with sodium hydroxide does not lead to the expected mercury(I) ox­ ide, but rather to a mixture of finely divided elemental mercury and mercury(II) oxide. The chalconides of divalent mercury exist in na­

of Extractive

Metallurgy

ture as minerals: the oxide HgO as montroydite, the sulfide HgS as cinnabar and metacinnabar, the selenide HgSe as tiemannite, and the telluride HgTe as coloradoite. They can also be produced synthetically. Only the oxide and the sulfide are of practical im­ portance. Mercury(II) oxide, HgO, ρ 11.1 g/cm3, is a red or yellow powder. The color depends on the size of the crystals: the yellow oxide con­ sists of crystals < 2 μm; the red one of crystals > 8 μm in diameter. Samples with particle sizes between these two values appear yellowto red-orange. Increasing the temperature leads to an intensification of the color: the yel­ low oxide becomes yellow-orange; the red one, dark red. The crystal lattice is rhombic and identical for both forms. Under certain preparative conditions a hexagonal form ex­ ists. This modification has no practical impor­ tance and can be converted to the more stable rhombic form by heating above 200 °C. Heating above 450 °C causes the oxide to decompose into elemental mercury and oxy­ gen. Mercury(II) oxide is sparingly soluble in water and in ethanol. With dilute mineral ac­ ids, solutions of the corresponding salts are formed, a method that can be used to prepare these salts. Production. Mercury(II) oxide can be pre­ pared via the anhydrous route by reaction of the elements at 350-420 °C under oxygen pressure or by thermal decomposition of mer­ cury nitrates at ca. 320 °C. Production via the wet route by precipitation is more important: the oxide is precipitated from solutions of mercury(II) salts by addition of caustic alkali [usually mercury(II) chloride solutions with sodium hydroxide]. Whether the yellow or the red form is obtained depends on reaction con­ ditions: slow crystal growth during heating of mercury with oxygen or during thermal de­ composition of mercury(I) nitrate leads to rel­ atively large crystals (i.e., the red form). Rapid precipitation from solution gives finer parti­ cles (i.e., the yellow form). Nevertheless, de­ pending on the conditions during precipitation such as stirring speed, pH, temperature, and

Mercury

method of mixing the components, large crys­ tals can be obtained by the wet route and, therefore, the red form is produced [101]. Uses. Red mercury(II) oxide in particular has become increasingly important for the production of galvanic cells with mercury ox­ ide anodes in combination with zinc or cad­ mium cathodes. These cells are distinguished from other systems in that their voltage re­ mains very constant during discharge; they are used mainly as small button-shaped batteries, e.g., for hearing devices, digital watches, ex­ posure meters, pocket calculators, and secu­ rity installations. Additional uses of mercury(II) oxide include the following: for the production of mercury(II) salts by treat­ ment with the corresponding acids, and as a reagent in analytical chemistry. Its impor­ tance as an additive to antifouling paint for ships and in medicine (e.g., for eye ointment) has decreased. Mercury(II) sulfide, HgS, is the most impor­ tant starting material for mercury extraction; it can exist in two forms: α-HgS (cinnabar, cinnabarite) has a density of 8.1 g/cm3, and (βHgS (metacinnabar) has a density of 7.7 g/cm3. The β-form slowly changes to α-HgS on heating. The latter sublimes at 583 °C. Of the two sulfide minerals, cinnabar is the most important ore for the production of mercury (see Section 17.4.1.2). When pure, the com­ pound is bright red and forms hexagonal crys­ tals. Metacinnabar is black and forms cubic crystals (zinc blende lattice). Both have ex­ tremely low water solubility; they are also in­ soluble in mineral acids and in caustic alkali. They dissolve only in aqua regia, to release sulfur, and in alkali sulfide solutions, to form thio complex-salt ions, such as [HgS 2 ] 2 - . Production by either the dry or the wet route is possible. In the former, a mixture of mercury and sulfur is heated. The elements re­ act slowly together even on mixing. Produc­ tion from aqueous solutions is more important. The sulfide is precipitated from so­ lutions of mercury(II) salts by treatment with hydrogen sulfide, alkali, or ammonium sulfide solutions. Initially, the black sulfide is formed.

911

It can be converted to the more stable red form by heating in the presence of the mother liquor or with ammonium polysulfide solutions. The reaction of mercury and sulfur by heating with a solution of sodium polysulfide has been de­ scribed [102].

17.11.2

Mercury Halides

Halides of both mono- and divalent mer­ cury are known. Of these, only the fluorides are ionic compounds: they undergo hydrolysis with water; mercury(I) fluoride simulta­ neously undergoes disproportionation. The other halides either are already composed of molecules in the crystal lattice or form these by dissolution or evaporation. Accordingly, their melting and boiling points are low. The halides of monovalent mercury are sparingly soluble in water: the solubility of divalent mercury halides decreases with increasing molecular mass. Mercury halides form numer­ ous basic compounds. Mercury(II) fluoride, HgF 2 , mp 645 °C, ρ 9.0 g/cm3, forms colorless octahedral crystals with a cubic ionic lattice (fluorite type). It is unstable in humid air; hydrolysis yields a yel­ low color. Mercury(II) fluoride is insoluble in organic solvents. The compound is produced from mercury and fluorine at elevated temperature or from mercury(II) oxide and hydrogen fluoride un­ der oxygen pressure at 450 °C [103]. Synthe­ sis from mercury(II) oxide and sulfur tetrafluoride has been suggested [104]. The compound is used in organic synthesis as a fluorinating agent. Mercury(I) chloride, calomel, Hg 2 Cl 2 , ρ 7.15 g/cm3, is rarely found as a mineral in nature. When pure, it exists as a heavy white powder or as colorless crystals with a silvery luster, having a tetragonal molecular lattice. It sub­ limes at 385 °C; above 400 °C the molecules decompose into a vapor composed of mercury and mercury(II) chloride. The substance is sparingly soluble in water, ethanol, diethyl ether, and acetone. A black color occurs in ammonia solution, whereby a mixture of

912

finely divided elemental mercury and mercury(II) ammonium chloride is formed by disproportionation—hence, the name calomel, from the Greek word meaning beautiful black. Production. An intimate mixture of mer­ cury and mercury(II) chloride is heated at 525 °C in closed iron or fused silica tubes, at­ tached to cooled receivers in which calomel vapor condenses [105, 106]. Synthesis from the elements is also possible [101]. Very finely divided mercury(I) chloride can be obtained by precipitation from a dilute nitric acid solu­ tion of mercury(I) nitrate and sodium chloride [105]. Mercury(I) chloride finds application in calomel electrodes, which serve as standard electrodes for the measurement of electro­ chemical potential; it is also employed as a fungicide and insecticide in agriculture, and as a catalyst in organic synthesis. Mercury(I) chloride is mixed with gold for painting on porcelain. Mercury(II) chloride, corrosive sublimate, HgCl 2 , mp 280 °C, bp 303 °C, ρ 5.43 g/cm3, is a white, heavy, crystalline powder with a rhombic crystal lattice. At the melting point the vapor pressure is 560 kPa; the substance can, therefore, sublime under reduced pres­ sure. The sublimate is moderately soluble in cold water. Its solubility increases sharply with increasing temperature: the saturation limit is 6.2% at 20 °C and 36.0% at 100 °C; the compound may therefore be purified by recrystallization from water. The sublimate is readily soluble in organic solvents, in contrast to mercury(I) chloride, so that a clear solution, for example, in ether, is an indication of the absence of calomel. Molecules of mercury(II) chloride exist as such in all solvents; mixing Hg 2+ ions with Cl - in aqueous solution leads immediately to undissociated mercury(II) chloride molecules; this process is used ana­ lytically to bind chloride ions in determining the COD of effluents. Aqueous solutions of sublimate are weakly acidic, resulting from hydrolysis of a small amount of the chloride. The sublimate is much more soluble in alkali chloride solution than in pure water, because

Handbook

of Extractive

Metallurgy

of the formation of chloro complex ions, e.g., 2[HgCl4] . Production. For the formation of mercury(II) chloride from the elements, mercury is oxidized with chlorine in heated retorts; the reaction is carried out with the appearance of flame at > 300 °C. The escaping sublimate va­ por is condensed in cooled receivers, where it settles as fine crystals. Formation of mercury(I) chloride is avoided by the use of ex­ cess chlorine [101]. Mercury and chlorine also react in the presence of water; in this case, in­ tensive stirring is necessary. The chloride formed precipitates as crystals after the solu­ bility limit has been exceeded. If an alkali chloride solution is used in place of water, so­ lutions of chloro complex salts are formed, which are used mainly for the production of other compounds of divalent mercury [107]. Mercury(II) chloride can also be prepared from other mercury compounds. Mercury(II) sulfate, for example, is heated in the dry state with sodium chloride, and the evolving mercury(II) chloride vapor is condensed to a solid in receivers. A warm sublimate solution is ob­ tained from the reaction of mercury(II) oxide and a stoichiometric amount of hydrochloric acid; the chloride separates as crystals on cooling. Uses. Mercury(II) chloride is an important intermediate in the production of other mer­ cury compounds, e.g., mercury (I) chloride, mercury(II) oxide, mercury(II) iodide, mercury(II) ammonium chloride, and organic mercury compounds. The compound is used as a catalyst in the synthesis of vinyl chloride, as a depolarizer in dry batteries, and as a re­ agent in analytical chemistry. It has minor im­ portance as a wood preservative and retains some importance as a fungicide. Other uses (e.g., as a pesticide or in seed treatment) have declined considerably. Mercury(II) bromide, HgBr 2 , mp 236 °C, bp 320 °C, ρ 6.05 g/cm3, forms colorless crystals with a rhombic layered lattice. Its water solu­ bility is highly temperature dependent: the concentration limit is 0.6% at 25 °C and 18% at 100 °C. Mercury(II) bromide is readily sol-

Mercury

913

uble in diethyl ether and ethanol. The com­ pound is produced from mercury and bromine in the presence of water; by dissolution of mercury(II) oxide in hydrobromic acid; or by precipitation from a nitric acid solution of mercury(II) nitrate with addition of sodium bromide [108]. It is used as a reagent for ar­ senic and antimony, as an intermediate in the production of bromine-containing organomercury compounds (see Section 17.12), and as a catalyst in organic synthesis. The melt is used as a nonaqueous solvent. Mercury(II) iodide, HgI 2 , exists in a red and a yellow form. Red mercury(II) iodide is solu­ ble in diethyl ether, chloroform, and methanol. The compound is produced by adding an aqueous solution of potassium iodide to an aqueous solution of mercury(II) chloride with stirring; the precipitate is then filtered off, washed, and dried at 70 °C. A yellow product is formed by sublimation; on cooling, it turns red. Yellow mercury(II) iodide, mp 259 °C, bp 354 °C, is unstable at room temperature and is converted to the red form at the slightest touch or upon heating. The transition temperature is 127 °C. Yellow mercury(II) iodide is prepared by pouring an alcoholic solution of the red form into cold water, whereby a pale yellow emulsion is formed, from which the mercury(II) iodide crystallizes after a few hours. Potassium mercury iodide, K 2 HgI 4 , readily soluble in water, is prepared by dissolving mercury(II) iodide in a concentrated potas­ sium iodide solution.

17.11.3

Mercury Pseudohalides

The cyanides and thiocyanates of divalent mercury resemble the halides, in that they also exist in solution as undissociated molecules and form highly soluble complexes with an excess of the anion; moreover, numerous basic compounds are derived from them. Mercury(I) cyanide is unknown; the poorly solu­ ble mercury(I) thiocyanate, which can be obtained by precipitation from mercury(I) ni­ trate solution, has no practical importance.

Mercury(II) thiocyanate, Hg(SCN) 2 , ρ 3.71 3 g/cm , is a white powder which is thermally unstable. Decomposition begins at 110 °C and becomes spontaneous at 165 °C, with the compound swelling to many times its normal volume. In air a blue flame appears, and a dark-colored, voluminous residue is left. The thiocyanate is sparingly soluble in cold water. It is produced by precipitation from mercury(II) nitrate solution with a stoichiometric amount of potassium thiocyanate solution and is used as an analytical reagent and as an intensifier in photography.

17.11.4

Acetates, Nitrates, Sulfates

The acetates, nitrates, and sulfates of mer­ cury are composed of ions and undergo hy­ drolysis with water. The acetate and the sulfate of monovalent mercury are sparingly soluble and the nitrate is quite soluble in dilute acid. Compounds of divalent mercury are readily soluble in dilute acid. Mercury(II) acetate, (CH 3 COO) 2 Hg, ρ 3.27 g/cm3, mp 178 °C, exists as a fine white pow­ der or colorless shiny crystal flakes. It is solu­ ble in diethyl ether and ethanol. It is produced by dissolution of mercury(II) oxide in dilute acetic acid and concentration of the resulting solution. Mercury(II) acetate is used for the synthesis of organomercury compounds, as a catalyst in organic polymerization reactions, and as a reagent in analytical chemistry. .

Mercury(I) nitrate, Hg 2 (N0 3 ) 2 2H 2 O, ρ 4.68 g/cm3, mp 70 °C, which forms colorless crys­ tals (monoclinic ionic lattice), is produced by dissolving mercury in cold dilute nitric acid and crystallizing the compound from the re­ sulting solution. Uses. Mercury(I) nitrate is the most readily available soluble salt of monovalent mercury and is, therefore, an important intermediate for other mercury(I) derivatives; sparingly solu­ ble compounds can easily be prepared from it by precipitation from aqueous solution. Ther­ mal decomposition leads to red mercury(II) oxide. "Millons reagent", a solution of mer­ cury nitrate and nitrous acid in dilute nitric

914

acid, is used as an analytical reagent for the in­ dication of tyrosine-containing proteins. Mercury(II) nitrate, Hg(NO 3 ) 2 or . Hg(N0 3 ) 2 H 2 O, forms colorless, hygroscopic crystals. Apart from the anhydrous salt and the monohydrate, several other hydrates and basic compounds are known. Mercury(II) nitrate is produced by dissolving mercury in hot con­ centrated nitric acid; the resulting solution is concentrated and the nitrate crystallized by cooling. Uses include the production of other divalent mercury derivatives, as a nitrating agent in organic synthesis, and as an analytical reagent. Mercury(I) sulfate, Hg2SO4, ρ 7.56 g/cm3, is a colorless microcrystalline powder (monoclinic crystal system) that is very sensitive to light. The compound is prepared by precipita­ tion from a solution of mercury(I) nitrate with sulfuric acid or sodium sulfate solution, or by electrochemical oxidation of mercury in dilute sulfuric acid. It is used as a depolarizer in stan­ dard cells after Clark and Weston. Mercury(II) sulfate. H g S O4 , ρ 6.49 g/cm3, is a white powder (rhombic crystal type). It is prepared by fuming mercury with concen­ trated sulfuric acid or by dissolving mercury(II) oxide in dilute sulfuric acid and evaporating the resulting solution until the compound crystallizes. Mercury(II) sulfate is used in analytical chemistry to bind chloride ions in the determination of the COD of waste­ water [109]; as a catalyst in the production of acetaldehyde from acetylene and of anthraquinonesulfonic acids; and as a depolar­ izer in galvanic elements.

17.11.5 Mercury-Nitrogen Compounds Reaction of mercury(II) compounds with ammonia solution leads, depending on reac­ tion conditions, to amine complexes or mercury(II) nitrogen compounds. Of the numerous compounds known, only mercury(II) amidochloride has any practical im­ portance.

Handbook

of Extractive

Metallurgy

Mercury(II) amidochloride, HgNH 2 Cl, ρ 3 5.38 g/cm , is a fine, white crystalline powder with a rhombic crystal lattice. It is insoluble in water and in ethanol, and soluble in warm acid, ammonium carbonate solution, and so­ dium thiosulfate solution. On heating it de­ composes without melting. The compound can be precipitated from mercury(II) chloride solution with an ammonia solution. Mercury(II) amidochloride is used in the treatment of severe skin disease, as an eye ointment, and as a veterinary preparation; its importance is declining because of the development of mer­ cury-free products.

17.11.6 Analysis, Storage, and Transportation; Protective Measures Analysis. Determination of the purity of mer­ cury compounds, which consists of the deter­ mination of trace amounts of foreign cations and anions, involves specific reactions of these ions [110]. conventional methods for the trace analysis of cations have been supple­ mented or replaced by methods involving the simultaneous determination of several ele­ ments by means of plasma emission spectros­ copy [111]. Storage and Transportation. Plastic-lined steel drums are normally employed for pack­ ing. Small amounts (e.g., for chemical labora­ tories) are usually placed in plastic or glass bottles. Many compounds are light-sensitive and must, therefore, be adequately protected from light sources. Proper consideration must be given to the chemical and toxic properties of mercury compounds and the necessary pro­ tective measures; this is particularly true in container labeling. Protective Measures. Most mercury com­ pounds, because of their toxicity, require the same protective measures during production and processing as metallic mercury. Further­ more, environmental protection necessitates appropriate precautions (see Section 17.5).

Mercury

915

17.12 Selected Organic Compounds Organic mercury derivatives [112, 113] are among the oldest known organometallic compounds. Organometallic compounds of divalent mercury are stable toward air, oxygen, and water. In contrast, organic derivatives of monovalent mercury are unstable and can be prepared only at low temperature. After 1990, use of organic mercury derivatives in Germany will be difficult because mercury will not be allowed in industrial effluent or wastewater. Organomercury compounds can be divided into two major groups: compounds of the type R - H g - R , where R and R are aliphatic or aromatic groups, and compounds of the type R-Hg-X, where R is aliphatic or aromatic and X is a halogen or an acid group. Synthesis of organic mercury compounds can be carried out by reaction of Grignard reagents with mercury halides [114, 115]. To obtain pure products, the mercury salt and the Grignard reagent must contain the same anion 1

2

1

RMgX + HgX

2

RHgX + MgX

2

2

Furthermore, organic mercury compounds can be produced by the reaction of sulfuric acids or their sodium salts with mercury(TI) halides [116]: R-SO H + HgX 2

R-HgX + SO + HX

2

2

Hydroxy- or alkoxymercury derivatives can be obtained via the solvomercuration reaction (Hofmann-Sand reaction) by addition of mercury(II) compounds to alkenes in aqueous, alkaline, or alcoholic solutions [114, 117]: 1

2

3

R - C H = C H - R + Hg(OAc) + R O H 2

3

1

2

(R O)CHR -CHR HgOAc + AcOH

Organic mercury compounds can be converted to other organometallic derivatives by electrolysis or transmetallization [118]. Dialkyl- or Alkoxyalkylmercury Compounds. Dimethylmercury, C H - H g - C H , a colorless, sweet-smelling liquid, is very toxic; it has a density of 3.069 g/cm , bp 96 °C, and is soluble in ethanol and diethyl ether. In acidic aqueous solution, dialkylmercury compounds are hydrolyzed to monoalkylmercury deriva3

3

3

tives. Dimethylmercury is formed in organisms by enzymatic methylation of elementary mercury or a mercury compound. In an organism it is easily converted to methylmercury ( C H - H g ) or methylmercury(II) chloride (CH -Hg-Cl). These species can react with free S-H groups of biologically important molecules, leading to the diseases that have been caused by mercury and its derivatives. Mixed dialkylmercury compounds (e.g., methylpropylmercury compounds) are less volatile than symmetrical dialkylmercury compounds (e.g., diethylmercury). Dialkylmercury derivatives are generally very reactive and can also undergo transalkylation with simple alkyl halides (e.g., ethyl iodide). +

3

3

Diarylmercury Derivatives. The best known diarylmercury derivative is diphenylmercury, (C H ) Hg, which is produced by direct mercuration. 6

5

2

Analysis. Organic mercury compounds may be analyzed in the following ways: they can be determined qualitatively (1) by digestion with concentrated sulfuric acid and 30% hydrogen peroxide or (2) by digestion with 10% sulfuric acid and subsequent addition of dithizone (diphenylthiocarbazone) in carbon tetrachloride solution. In the presence of mercury the green solution turns orange. Organic mercury compounds can be determined quantitatively by atomic absorption spectroscopy.

17.13

Economic Aspects

A detailed description of the development of production is given in the metal statistics of Metallgesellschaft AG, Frankfurt [119] (Table 17.12). Over the last ten years, production figures have changed only slightly. According to U.S. estimates, current production is ca. 53% of the potential capacity. Because of reduced demand many mines and smelting plants are no longer operating or have greatly cut back production. Intervention by the former Soviet Union has seriously depressed the price of mercury to dumping price levels, although this has been resisted by Spain and Algeria. The change in the price of mercury

Handbook

916

since 1979 is shown in Table 17.13 [119]. A survey of previous mercury prices is included in [120]. The main producers of mercury extracted by mining are as follows [ 121 ] (figures refer to flasks): Algeria China Finland Yugoslavia Mexico Spain Former Czechoslovakia Turkey Former Soviet Union United States

23 20 2 2 10 42 4 6 66 14

000 000 300 000 000 000 400 000 000 000

Metallurgy

350 t of mercury (ca. 10 000 flasks) was extracted from mines in 1986, a large part of the production being exported to Brazil and Argentina. Spanish mercury production is ca. 40 000 fl. The main importers are the United States, Belgium, Luxemburg, and France. The former Soviet Union produces ca. 67 000 fl of mercury; in contrast to the high export level in 1986, almost the entire production is now reserved for domestic use. The capacity of production plants at the Anzob antimony mercury complex in Tajikistan has been doubled. The major proportion of former Yugoslav production of ca. 2000 fl of mercury is exported.

1

The Algerian mercury producer, Entreprise nationale des non-ferreux et substances utiles (ENOF), quotes production prices of $300 per flask. Most of the Chinese production is exported to the United States. China claims to have the largest mercury resources in the world. Guizhou Province contains five mines, accounting for ca. 90% of Chinese production. Italy, once a large mercury producer, now imports mercury from Algeria, the Netherlands, and former Yugoslavia. In Mexico, ca. 1

of Extractive

The sole producer in the United States discontinued mining operations from the middle of 1987 to February 1988. The production for 1988 is 14 000 flasks.

The development of prices and production figures will certainly be influenced substantially by acceptance of the metal mercury, and in large sectors no foreseeable substitute exists. The utilization of mercury in a highly industrialized country such as Germany is shown in Table 17.14 [122]. The purchase of mercury by individual sectors has decreased sharply since 1980. The main user of mercury is the electrical engineering industry, followed by the alkali chloride industry. The use of mercury in paints, pigments, and pesticides has fallen sharply.

Table 17.12: Production of mercury in tonnes [119]. 1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1672 1520 80 72

1757 1539 130 88

1693 1471 147 75

1764 1553 144 67

1704 1499 135 70

1178 967 160 51

1140 962 141 37

135 52 74

128 36 85 7

734 636 98

475 386 89

Asia (Turkey)

182

226

262

211

97

202

60

25

5

Africa (Algeria)

586

801

764

756

662

587

639

431

476

459

400

America Dominican Republic Mexico United States

1043 2 384 657

965 1 394 570

695

264

724

1065

1195

398

85

82

80

185 510

124 140

345 379

651 414

735 460

340 58

12 70

10 70

Total Western countries Czechoslovakia USSR China

3483 152 1220 800

3749 158 1200 800

3414 168 1200 850

2995 164 1200 900

3187 168 1180 900

3032 131 1180 1200

3034 126 2100 800

989 75 1900 780

694 60 1900 392

1275 50 1700 468

955 50 1500 408

Total Eastern countries

2172

2158

2218

2264

2248

2511

3026

2218

1958

5655

5907

5632

5259

5435

5543

6060

2755 3744

2352

Total world

3046

3493

2913

a

Europe Spain Finland Yugoslavia a

b

b

a

b

Excluding Estimates.

Eastern-Bloc

countries.

b

b

b

b

b

9

b

b

b

b

21 64

b

b

Mercury

917

The toxicity of mercury is based upon its action as a general cell and protoplasmic poison, i.e., bonding to the sulfhydryl groups of proteins; denaturing proteins; damaging membranes; and reducing the RNA content of cells. This leads to blocking of many enzyme systems. The kidneys and nervous system are especially vulnerable. In animal experiments, methyl mercury and mercury(II) chloride cause a dose-dependent suppression of spermatogenesis.

Table 17.13: Quoted prices for mercury in London and New York. European price Year

£ per flask (34.473 kg)

1985 1986 1987 1988 1989 1990 1991 1992 1993 1994

284.02-293.24 187.49-200.36 245.64-255.23 297.17-310.49 246.43-261.63 200.91-218.79 103.33-119.44 126.85-173.35 107.02-125.05 103.23-119.32

U.S.

£/kg

Price

$ per flask (34.473 kg) $/kg

8.24-8.51 5.44-5.81 7.13-7.40 8.62-9.01 7.15-7.59 5.83-7.22 3.00-3.46 3.68-5.03 3.10-3.63 2.99-3.46

310.957 232.785 295.503 335.517 287.722 249.218 122.424 201.390 186.510 194.453

9.02 6.75 8.57 9.73 8.35 7.23 3.55 5.84 5.41 5.64

Acute poisoning occurs when mercury ion concentrations reach 0.2 mg per 100 mL of blood. Daily 5-h exposure to inhaled mercury vapor concentrations of 0.1 mg/m leads to severe chronic mercury poisoning. 3

17.14 Toxicology and Occupational Health [123]

In foods, different mercury concentrations are tolerated: e.g., in the United States, 0.05 ppm; in Germany, 0.1 ppm. Environmental contamination with mercury leads to a critical concentration effect in animals that occupy higher positions in the food chain (large fish and fish-eating sea fowl). In certain fish, concentrations of 10 ppm of mercury and more have been found: fish and clams originating in Japan's Minamata Bay contained up to 9.6 mg of mercury per kilogram. Eggs of wild birds on the Finnish coast contained up to 3.5 mg of mercury per kilogram. Up to 270 mg of mercury per kilogram of organ tissue was found in dead seed-eating birds.

Uptake, Mode of Action, Metabolism. The toxicity of mercury depends, among other things, on its state of aggregation and degree of dispersion. Both fine particulate dust containing mercury and mercury vapors are very toxic in comparison to the liquid metal. Various mercury compounds are very potent poisons. Compounds containing divalent mercury are generally more poisonous than monovalent ones. The toxicity of inorganic mercury compounds increases with increasing solubility. Still, in most cases, they are less toxic than organic mercury compounds.

Table 17.14: Breakdown of mercury use (in tonnes) in former West Germany according to sector [122]. 1973 Alkali chloride industry (without changes in stock levels)

1974

128.0 127.0

1975

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

99.0 103.0

78.0

87.5

87.5

72.0

50.4

46.3

40.7

31.6

41.8

27.5

18.1

14.1

17.9

6.0

4.5

3.9

9.1

3.8

3.5

3.8

1.0

0.6

0.3

Catalysis

45.0

29.0

9.0

10.0

11.6

4.8

Paints, dyes

19.5

18.9

5.6

12.4

12.6

12.6

Pesticides

30.5

33.7

26.6

27.6

28.7

28.5

33.0

31.8

26.7

4.1

19.0

14.0

9.0

Electrical engineering

26.2

42.1

40.6

43.4

40.4

43.5

45.4

51.5

55.0

54.2

53.1

56.2

65.8

Control instruments and apparatus construction

14.8

17.3

8.7

10.3

14.6

13.4

13.3

16.5

15.7

17.1

7.0

6.1

7.2

Chemicals and reagents

25.3

7.0

33.1

45.8

22.0

31.4

43.2

35.5

27.1

53.9

36.2

24.7

Medicine

24.2

25.2

25.0

26.0

23.9

24.3

24.2

24.2

22.9

23.6

23.5

24.3

24.1

Miscellaneous

32.8

31.1

44.2

46.7

30.0

30.0

30.0

42.1

32.5

35.7

30.1

30.2

30.2

Total Total, % (1980= 100%)

346.3 331.3 291.8 325.2 261.8 276.0 313.2 295.5 247.9 256.6 216.6 192.2 182.3 117

112

99

110

89

94

106

100

84

86

73

65

62

918

Humans are estimated to consume 0.2 mg of mercury weekly in their diet. Mercury uptake with food leads to concentrations in the kidney of < 0.1-3 mg/kg; the corresponding concentrations with intoxication are 10-70 mg/kg. Amalgam fillings are the most frequent cause of chronic mercury and tin intoxication. The metal is transformed to highly toxic organic compounds by oral microorganisms. During chewing (gum) up to 26 mg/L of mercury and 0.3 mg/L tin may be set free. After 15 min, the mercury content of the respiratory air increased eightfold. Toxicity of Metallic Mercury and Inorganic Mercury Compounds. The toxicity of mercury and its compounds depends predominantly on their solubility, which determines absorption and distribution in the organism. Thus metallic mercury and all mercury compounds are toxic, with the exception of red mercury sulfide, which is practically insoluble in the body. Mercury vapor causes acute damage to the lungs and chronic damage to the central nervous system. Mercury salts are caustic to the mucous membranes of the gastrointestinal tract and nephrotoxic when absorbed. Metallic Mercury. Liquid mercury is not highly toxic; in earlier times, it was used as a treatment for ileus. Thermometers that break in the rectum lead to mercury intoxication only if a wound is created and mercury is pressed into the tissues from which it is slowly absorbed. Granulomas develop when mercury is injected into subcutaneous fat tissue; this can occur in suicide attempts or injuries to the hand caused by broken thermometers. Absorption occurs when the mercury depot is broken up into minuscule droplets, for example, after surgical excision. Occasionally liquid mercury has been injected intravenously, either suicidally [124, 125] or accidentally during intracardiac catheter studies when blood was drawn for blood gas analysis by using mercury-filled syringes. Mercury embolisms resulted, and some of the patients showed symptoms of intoxication. One of

Handbook

of Extractive

Metallurgy

nine patients died after five months as a result of the intoxication. Blindness as a result of occlusion of the central artery also occurred. Antisyphilitic treatment with gray mercury ointment caused numerous intoxications, with all grades of severity being encountered. They resulted from skin absorption and from inhalation of the mercury that vaporized on the skin. Acute inhalant poisoning is very dangerous. In four cases, after several hour's exposure to mercury vapor concentrations of 1-3 mg/m , acute pneumonitis resulted. Chronic inhalant intoxication can be expected with mercury vapor concentrations of 0.1-1 mg/m . With < 0.1 mg of mercury per cubic meter, even mild intoxication is improbable. In sensitive persons, an increase in subjective signs (micromercurialism) has been observed at 0.02-0.1 mg/m [125]. Mercury Salts. Salts of divalent mercury are more toxic than monovalent ones, regardless of the route of administration. In animal experiments the LD after parenteral injection for divalent salts is ca. 5 mg of mercury per kilogram. As a result of their poor absorption, they are much less toxic if administered orally, the LD in this case being of the order of 100 mg of mercury per kilogram [126]. Strongly dissociated salts are more caustic and generally more toxic than less dissociated ones. An exception to this is mercury oxycyanide [Hg(CN) HgO], which is highly poisonous even though it hardly dissociates. Cyanide ions may enhance the toxicity. Acute Poisoning in Humans. Mercury(II) chloride (corrosive sublimate) is one of the strongest corrosive poisons; for adults, oral doses of 0.50-1.0 g (in several cases even 0.2 g) are fatal, even though people have survived after ingesting 5 g. A total of 0.2 g introduced into the vagina can be fatal. Administration of 1.5 g of mercury oxycyanide was lethal. The toxicity of mercury(I) chloride (calomel) depends on its retention time in the gastrointestinal tract. The lethal dosage for adults is generally 2-3 g; for children, 0.4 g [125]. However, in former times, therapeutically administered doses of 0.1 g of mercury(I) chloride have led to death, especially when the 3

3

3

50

50

.

2

Mercury

laxative effect did not occur. Chronic intoxica­ tion with mercury salts is unusual; in indus­ trial poisonings, exposition to mercury vapors generally exists concomitantly. Mercury salts act as direct skin irritants. Furthermore, they are sensitizing, especially mercury fulminate. Toxicity of Organic Mercury Compounds. Most organic mercury compounds are lipid soluble. Some of them vaporize easily and thus also act in the gaseous phase. Organic mercury compounds can cause toxic dermati­ tis and, as a result of their lipid solubility, can severely damage the central nervous system. Hypersensitivity reactions, as well as kidney damage, also occur. In the organism, phenylmercury and alkoxyalkylmercury compounds are metabo­ lized to inorganic mercury compounds and act like mercury salts. The stable alkylmercury compounds are neurotoxic and embryotoxic. Alkylmercury Compounds. Numerous tox­ icity studies on animals have been reported. The LD 50 in rats and mice of most methyl- and ethylmercury compounds is 10-30 mg of mer­ cury per kilogram. This holds true for parenteral and oral administration [126]. In humans, almost all alkylmercury poisoning has been caused by contaminated food, almost all involved chronic poisoning, the dose in­ gested is unknown, and estimated values are questionable. Mass poisoning in Iraq was caused by pita bread baked with flour made from seed grain; the grain contained ca. 15 mg of mercury per kilogram of alkylmercury (mainly methylmercury) salt. The bread weighed 220 g and had a water content of 31 %. Affected adults are estimated to have eaten six to eight loaves per day; the first fatalities occurred six to eight weeks after distribution of the seed grain [126, 127]. Minamata disease developed in fishermen and their family members who ate fish daily or at least several times a week, consuming 250500 g of fish with each meal. The average me­ thylmercury concentration of the fish eaten is not known (estimates: 5-20 mg of mercury per kilogram of fish). Thus, a daily uptake of 1.5-4 mg of mercury would have resulted, re­

919

spectively [4]. Affected patients had mercury concentrations of 200-2000 μg/L in their blood and 50-500 mg/kg in their hair [128]; the brain of patients who died contained > 5 mg of mercury per kilogram [126]. Mild symptoms are assumed to occur with concen­ trations of 100 μg of mercury per liter of blood and 30 mg of mercury per kilogram of hair. Daily ingestion of 5 μg of mercury per kilo­ gram of body weight in the form of methylmercury compounds is considered the minimal toxic dose [128]. Mercury is mutagenic, teratogenic, and em­ bryotoxic, especially in the form of alkylmer­ cury compounds [129, 130]. The fetus is three to four times more sensitive to methylmercury than the pregnant woman [128]. Congenital brain damage occurred in 5-6% of the chil­ dren from Minamata Bay, where the rate ex­ pected was 0.1-0.6%. The mothers belonged to the group of people heavily exposed to me­ thylmercury but did not show any clear symp­ toms of intoxication [126]. Arylmercury Compounds. In animal experi­ ments, arylmercury compounds are as toxic as alkylmercury compounds when administered parenterally; administered orally, however, they are less toxic. As a result of their instabil­ ity they act like a combination of organically bound mercury and mercury vapor. In adults, ingestion of 100 mg of mercury in the form of phenylmercury nitrate led to abdominal pain and mild diarrhea; however, 120 mg was also tolerated without symptoms. Even after inges­ tion of 1.25 g of mercury, clinical chemistry values and kidney biopsy results were normal. Concentrations > 0.6 g/L are locally caustic [126]. Antagonism. Mercury and selenium are an­ tagonists; within certain limits, selenium can reduce the toxicity of inorganic and methyl­ mercury. In animal experiments, selenium pre­ dominantly delays the appearance of intoxication symptoms but reduces the lethal­ ity only slightly. Whether a high dietary sele­ nium content can protect against methylmercury poisoning is still open to ques­ tion: the fish that caused Minamata disease

920

Handbook

contained, in addition to large amounts of mercury, a high proportion of selenium [128, 131]. Occupational Health. The MAK value for metallic mercury in Germany is presently 0.1 mg/m (0.01 ppm); for mercury vapor the MAK value is 0.05 mg/m [123]. The TWA of mercury is 0.05 mg/m . The theoretically possible vapor pressure concentration can, however, far exceed this value. The odor threshold for mercury is 13 mg/m . The legal requirements covering industrial safety and hygiene when working with mercury and its compounds depend on the laws of individual countries. The measures adopted by a highly industrialized country such as Germany are described below. The special safety measures for handling and working with mercury-containing materials are given in the TRgA (Technische Regeln für gefährliche Arbeitsstoffe; technical regulations for dangerous substances). They do not cover mercury(II) sulfide, inorganic compounds containing < 0.1 % mercury, or organic compounds containing < 0.05 % mercury. The working methods and procedures must basically be designed so that employees are not exposed to mercury vapor, mist, or dust. Further details are given in [132]. A series of specification sheets for handling mercury has been published, which describes the technical and personal safety measures to be adopted [133]. Explanatory information and instructions for work and health safety are summarized in a comprehensive poisons list [134]. Persons working with mercury should be monitored regularly. With metallic mercury, inorganic mercury compounds, and organic nonalkyl mercury compounds, mercury values in urine should not exceed 150 μg/L. The blood levels should be < 35 μg/L. With or­ ganic alkylmercury compounds, the limiting blood level is 75 μg/L. 3

3

3

3

17.15 References 1. P. T. Craddock, Bull. Metals Museum 10 (1985) no. 10, 3-25.

of Extractive

Metallurgy

2. G. Agricola: 12 Bücher des Bergbaus und des Hüttenwesens, pp. 370-375. 3. R. C. Weast, M. J. Astle (eds.): Handbook of Chemistry and Physics, 64th ed., CRC Press Boca Raton, FL, 1983. 4. Hüls, DE 3709570, 1987 (S. Sridhar). 5. K. R. Suttill, Eng. Min. J. (1989) 2 4 - 2 7 . 6. W. Peters, Erzmetall 35 (1982) no. 7/8, 389-395. 7. H. Köse, M. Kemal, F. Simsir, Erzmetall 42 (1989) no. 6, 276-278. 8. SRI International, Chemical Economics Handbook, MenloPark 1984. 9. X. Wang, Z. Wu: Mineral Processing and Extractive Metallurgy, The Institution of Mining and Metallurgy, Portland Place, London 1984, p. 44. 10. L. D. Skrylio, L. M. Lopatenko, L. A. Sin'kova, Sov. Non-Ferrous Met. Res. (Engl. Transl.) 13 (1985) no. 3, 186-187. 11. V. Tafel: Lehrbuch der Metallhüttenkunde 2nd ed., vol. 1, Hirzel, Leipzig 1951, p. 590 ff. 12. F Pawlek: Metallhüttenkunde, Berlin, Waller de Gruyter, 1983. 13. Lumalampan, DE 3243813, 1982 (A. Sikander). 14. Linden Chemical & Plastics, EP 0042509, 1982 (R. J. Burkett). 15. Elektro-Ofenbau Matthias Marcus, DE 3609517, 1986. 16. H. Kulander, Sprechsaal 119 (1986) no. 11, 10161018. 17. F. Hiller: Entsorgung von Gerätebatterien, lecture held at the Technische Akademie Eßlingen, März 1988. 18. N. Hirayama: Behandlung von Sonderabfall 1, EFVerlag, Berlin 1987. 19. H. Pietsch, Bundesministerium für Forschung und Technologie, Forschungsprojekt FKZ: 143 03 553, Lurgi, Frankfurt 1986. 20. H. Laig-Hoerstebrock, Bundesministerium für Forschung und Technologie, Forschungsprojekt FKZ: 143 0273/0, Varta, Kelkheim 1986. 2 1 . H. Kulander: Behandlung von Sonderabfall 1, EFVerlag, Berlin 1987. 22. Preussag AG Metall, internal report, 1989, Goslar, Germany. 23. Outokumpu O.Y., DE 2406119, 1973. 24. Preussag AG Metall, DE 2558115, 1975 (G. Heyer). 25. K. Hanusch, BMFT-Forschungsbericht NTS 0103/0, Nov. 1973, Harlingerode. 26. J. L. Gallant, World Conf. Int. Nucl. Targ. Development Soc, Boston, MA, 13 Oct. 1979, Plenum Press 8(181), pp. 213-216. F. Desmet, L. Lemattre, A. P. van Petegham, Mater. Chem. Phys. 11 (1984) 305-309. 27. A. D. Pogorelyi, G. M. Tysh, Tsvetn. Met. 1979, no. 5, 10-18. 28. S. A. N. Sheya, J. H. Maysilles, R. G. Sandberg, Rep. Invest. U.S. Bur. Mines 9191 (1988) 1-12. 29. I. Barin, O. Knacke: Thermochemical Properties of Inorganic Substances, Springer Verlag, Berlin, Verlag Stahleisen, Düsseldorf 1973, p. 339. 30. Ch. G. Maier, Am. Inst. Min. Metall. Eng. Techn. Puhl. no. 264(1929). 3 1 . R. Heindryckx et al., "Mercury and Cadmium in Belgian Aerosols", in: CEC 1974 Problems of the Contamination of Man and his Environment by

Mercury

32.

33.

34. 35. 36. 37. 38.

39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51.

52. 53.

54. 55. 56.

57. 58.

Mercury and Cadmium, Commission of the European Communities, Luxemburg, pp. 135-148. N A S , An Assessment of Mercury in the Environment, National Academy of Sciences, Washington, DC, 1978. J. K. Piotrowski, D. O. Coleman: MARC-Report 20, Environmental Hazards of Heavy Metals: Summary Evaluation of Lead, Cadmium, and Mercury, MARC-Monitoring and Assessment Research Center, GEMS Global Environmental Monitoring System, Geneve, London 1980. H. J. Rösler, H. Lange: Geochemische Tabellen, 2nd ed., Enke Verlag, Stuttgart 1976. Schwermetalle in Lebewesen und Böden ANS-Mitteilungen, Sonderheft 2, p. 18 ff. P. Koronowski: Nebenwirkung von Quecksilberverbindungen auf Mensch und Tier, Berlin 1973. S. Bombach, L. Peters, Naturwissenschaften 62 (1975)575-576. E. Heinrich, H. Paucke, H.-D. Nagel, D. Hansen: Agrochemikalien in der Umwelt, VEB Fischer Verlag, Jena 1976, p. 57. R. D. Rogers, J. C. McFarlane, J. Environ. Qual. 8 (1979) no. 2, 255-260. R. Taylor, T. Bogacka, M. Balcerska, Environ. Prot. Eng. 4(1978) no. 2, 179-182. J. Lag, E. Steiners, Acta Agric. Scand. 28 (1978) 393-396. R. W. Klusman, R. A. Landrers, J. Volcanol. Geotherm. Res. 5 (1979) 4 9 - 6 5 . R. Bargagli, F. P. Iosco, C. Barghigiani, Water Air Soil Pollut. 36 (1987) 219-225. T. Stijve, R. Roschnik, Min. Geb. Lebensmittelunters. Hyg. 65 (1974) 208-220. R. Seegers, Z. Lebensm. Unters. Forsch. 160 (1976) 303-312. K. Aichberger, O. Horak, Bodenkultur 26 (1975) no. 1,8-14. W. Rauter, Z. Lebensm. Unters. Forsch. 159 (1975) 149-151. E. Hauser, M. Mohadjerani, Mitt. Geb. Lebensmittelunters. Hyg. 67 (1976) 3 8 9 - 4 0 1 . M. Cember, M. Curtis, B. G. Blaylock, Environ. Pollut. 17 (1978) no. 4, 311 -319. A. Stock, Z Angew. Chem. 41 (1928) 663. Blei-, Kadmium- und Quecksilbergehalte von Lebensmitteln in der Bundesrepublik Deutschland, Bericht der Zentralen Erfassungsund Bewertungsstelle für Umweltchemikalien im Bundesgesundheitsamt Berlin (ZEBS), Stand 1975, und ZEBS-Berichte 1/79. Arsen, Blei, Cadmium und Quecksilber in und auf Lebensmitteln, ZEBS-Bericht, Berlin, 1/1984. Quecksilber-, Magnesium- und Zinkgehalte in der Frauenmilch, im Blutserum und Fettgewebe der Mütter, ZEBS-Hefte, Berlin, 1/1986. A. Kuivala, J. Poijärvi, Erzmetall 30 (1977) 556. Boliden AB, EP 179040, 1985 (F. Dyvik). F. Dyvik: "Extraction Metallurgy 85", Proc. Conf. London 1985, The Institution of Mining and Metallurgy, London, pp. 189-198. Norddeutsche Affinerie AG, EP 278537, 1988 (H. Winkler, C. Reppenhagen). H. Braun, M. Metzger, H. Vogg, Müll. Abfall 18 (1986) no. 2, 6 2 - 7 1 , 8 9 - 9 5 .

921

59. Kernforschungszentrum Karlsruhe, DE 3715046, 1987 (H. Vogg, H. Braun, M. Metzger, A. Merz). 60. Kernforschungszentrum Karlsruhe, EP 0289810, 1988 (H. Vogg, H. Braun, M. Metzger, A. Merz). 61. Kernforschungszentrum Karlsruhe, EP 0289809, 1988 (H. Vogg, H. Braun). 62. Bergwerksverband, DE 3715526, 1987 (K. Knoblauch, K. Wybrands, K.-D. Henning, J. Degel, H. Ruppert). 63. D. Reimann, Umweltmagazin, Oct. 1984, 4 8 - 5 4 . 64. VEB Chemieanlagenbau- und Montagekombinat Leizig, DD 139068, 1976 (H. Tischendorf, R. Boege, G. Kreutzberger). 65. M. D. Rosenzweig, Chem. Eng. (London) 82 (1975, Jan. 20) 60-61; Chem. Eng. (London) 82 (1975, Feb. 3) 36-37. 66. Chlorine Inst.: Water Pollution Aspects, 5. 3. 1971. 67. K.-H. Bergk, F. Wolf, S. Eckert, Z. Chem. 17 (1977) 85-89. 68. G. van der Heiden, C. M. S. Raats, M. F. Boon, Chem. Ing. Tech. 51 (1979) 6 3 1 - 6 5 3 . 69. Kernforschungszentrum Karlsruhe, DE 3721141, 1987 (M. Metzger, H. Braun). 70. D. O. Reimann, VGB Kraftwerkstechnik 64 (1984) no 3,230-235. 7 1 . "Umwelt und Degussa, TMT-15 für die Abtrennung von Schwermetallen aus Abwässern", Degussa Broschüre, 1982. 72. Wacker-Chemie, DE 3335127, 1983 (J.-H. Janssen, B. Bangler, K.-H. Fahrmeier). 73. Hoechst AG, EP 0091043, 1983 (H. von Plessen, R. Gradl, G. Schimmel). 74. R. Kola, Erzmetall 30 (1977) no. 1 2 , 5 5 9 - 5 6 1 . 75. R. K. Williardson, A. C. Beer, Semicond. Semimetals 18 (1981) 21-45. 76. Montedison S.p.A., EP 0148023, 1984 (M. Gramondo, G. Donati, G. Faita, G. L. Marziano). 77. Chem. Eng. (London) 82 (1975) no. 2 , 6 0 . 78. Chem. Eng. (London) 82 (1975) no. 3, 36. 79. Chemikerausschuß der Ges. Dtsch. Metallhüttenund Bergleute: Analyse der Metalle, vol. 2, Springer Verlag, Berlin-Göttingen-Heidelberg 1953, p. 590. 80. Z. Anal. Chem. 96 (1934) 30. 8 1 . Z.Anal. Chem. 89 (1932) 187. 82. Z. Anal. Chem. 98 (1934) 3 3 1 . 83. R. Dumarey, R. Heindryckx, R. Dams, J. Hoste, Anal. Chim. Acta 107 (1979) 159-167. 84. G. R. Garnrick, W. Barnett, W. Slavin, Spectrochim. Acta Part B, 41 B (1986) no. 9, 991-997. 85. L. Sipos, J. Golimowski, P. Valenta, H. W. Nürnberg, Fresenius Z. Anal. Chem. 298 (1979) no. 1, 18. 86. H. Bloom, B. Noller: Trends in Electrochemistry, Badford, Australia 1976, 241-252. 87. H. Dorias: Gefährliche Güter, Eigenschaften, Handhabung, Lagerung und Beförderung, Springer Verlag, Berlin 1984. 88. G. Hommel: Handbuch der gefährlichen Güter, Springer Verlag, Berlin 1985. 89. J. M. Lo, C. M. Wai,Anal. Chem. 47 (1975) no. 11, 1869-1870. 90. S. Dogan, W. Herdi, Anal. Chim. Acta 101 (1978) no. 2, 433-136. 91. T. Suzuki, M. Fujita, K. Iwashima, Eisei Kagaku 26 (1980) no. 5, 229-235.

922

92. J. C. Meranger, B. R. Hollebone, G. A. Blanchette, J. Anal. Toxicol. 5 (1981) no. 1, 3 3 - 4 1 . 93. J. R. Kechtel, Analyst (London) 105 (1980) 8 2 6 829. 94. R. W Heiden, D. A. Aikens, Anal. Chem. 51 (1979) no. 1, 151-156. 95. P. R. Ludlam, J. G. King, Analyst (London) 106 (1981) 488-489. 96. R Schaaf, Müll Abfall 11 (1983) 277-283. 97. G. Jangg, H. Palman, Z. Metallk 54 (1963) 364. 98. G. Brauer, G. Düsing, Z. Anorg. Allg. Chemie 328 (1964) 154. 99. G. Jangg, Metall (Berlin) 13 (1959) 407. 100. C. Guminski, Z Metallk. 11 (1986) no. 2, 87-96. 101. N. P. Chopey, Chem. Eng. (NY.) 68 (1961) no. 25, 120. 102. A. Giordano, US 3061412, 1960. 103. Du Pont, US 2757070, 1956. 104. Du Pont, US 2904398, 1959. 105. R. A. Feldhoff, Pharm. Ztg. 15 (1930) 11. 106. R. Hirayama, JP 174101, 1946. 107. Wood-Ridge Chem. Co., US 3424552, 1967. 108. Y V. Karyankin, J. J. Angelov: Chisty Klimicheskie Reativy (pure chemical reagents), state-owned scientific-technical publisher for chemical literature, Moskva 1955, p. 457. 109. W. Leithe, Vom Wasser 37 (1970) 106. 110. Merck-Standards, pp. 807-830, 1971, Darmstadt, Germany. 111. J. Dahmen, Proc. 6th Indo-Ger. Semin. on Trace Element Analysis — Methods and Selected Applications, Oct. 1978 (Maria Laach). 112. L. G. Makarova, A. N. Nesemeyanow: The Organic Compounds of Mereury, North-Holland Pubi. Co., Amsterdam 1967. 113. Beilstein, E IV (4), 4426-4463; E IV (16), 1 7 0 1 1718. 114. C. Larock, Angew. Chem. 90 (1978)28-38. 115. C. Larock: Organo Mercury Compounds in Organic Synthesis, Springer Verlag, Berlin 1985. 116. W. Peters, Ber. Dtsch. Chem. Ges. 38 (1905) 2567. Bayer, DE 1003733, 1955. 117. J. B.Johnson, J. P. Fletcher, Anal. Chem. 31 (1959) 1563. 118. D. Steinborn, U. Sedlak, Z Chem. 25 (1985) 376 ff. 119. Metallstatistik 1987 der Metallgesellschaft AG, Frankfurt/Main 1987. 120. Ullmann, 4th ed., 19, 654.

Handbook

of Extractive

Metallurgy

121. United States Department of the Interior: Minerals Yearbook 1986, vol. 1, Bureau of Mines, US Government Printing Office, Washington 1988, p. 659 ff. 122. A. Rauhut, Metall (Berlin) 42 (1988) no. 11, 11371141. 123. D. Henschler (ed.): Gesundheitsschädliche Arbeitsstoffe. Toxikologisch-arbeitsmedizinische Begründung von MAK-Werten, Quecksilber, Verlag Chemie, Weinheim 1981. 124. H. E. Stockinger: " T h e Metals, Mercury, Hg", in: G. D. Clayton, F. E. Clayton (eds.): Patty's Industrial Hygiene and Toxicology, vol. 2A, John Wiley & Sons, New York-Chichester-Brisbane-Toronto 1981, 1769-1792. 125. E. W. Baader: "Quecksilbervergiftung", Handbuch der gesamten Arbeitsmedizin, vol. II/l, Urban & Schwarzenberg, München 1961. 126. L. Friberg: "Aspects of Chronic Poisoning with Mercury", Nord. Hyg. Tidskr. 32 (1951) 240-249, Arch. Ind. Hyg. 5 (1952) 596-597. L. Friberg, J. Vostal (eds.): Mercury in the Environment, CRC Press, Cleveland, OH 1974. 127. M. R. Greenwood: "Methylmercury Poisoning in Iraq. An Epidemiological Study in the 1971-1972 Outbreak", J. Appl. Toxicol. 5 (1985) 148-159. 128. M. J. Inskip, J. K. Piotrowski: "Review of the Health Effects of Methylmercury", J. Appl. Toxicol. 5 (1985) 113-133 129. J. O. Nriagu (ed.): The Biogeochemistry of Mercury in the Environment, Elsevier-North Holland Biomedical Press, Amsterdam-New York-Oxford 1979. 130. M. R. Greenwood: "Quecksilber", in E. Merian (ed.): Metalle in der Umwelt, Verlag Chemie, Weinheim 1984, 511-539. 131. E. J. Underwood: Trace Elements in Human and Animal Nutrition, Academic Press, New York-San Francisco-London 1977, 375-387. 132. Technische Regeln für gefährliche Arbeitsstoffe (TRgA), Berufsgenossenschaftliches Institut für Arbeitssicherheit 1983. 133. Berufsgenossenschaft der chemischen Industrie: Quecksilber und seine Verbindungen, Merkblatt M 0247/80, Jedermann-Verlag, Heidelberg 1980. 134. L. Roth, M. Daunderer: "Giftliste, Gifte, Krebserzeugende, gesundheitsschädliche und reizende Stoffe", Toxikologische Enzyklopädie, Ecomed, 36, Supplement 3/89.